The DOI matches its Crossref record; 60 of the 60 references it cites were found in a registry or an index; 11 of the 20 sources its claims cite were checked for support.
Expel: LLM agents are experiential learners
Open accessAbstract
The recent surge in research interest in applying large language models (LLMs) to decision-making tasks has flourished by leveraging the extensive world knowledge embedded in LLMs. While there is a growing demand to tailor LLMs for custom decision-making tasks, finetuning them for specific tasks is resource-intensive and may diminish the model's generalization capabilities. Moreover, state-of-the-art language models like GPT-4 and Claude are primarily accessible through API calls, with their parametric weights remaining proprietary and unavailable to the public. This scenario emphasizes the growing need for new methodologies that allow learning from agent experiences without requiring parametric updates. To address these problems, we introduce the Experiential Learning (ExpeL) agent. Our agent autonomously gathers experiences and extracts knowledge using natural language from a collection of training tasks. At inference, the agent recalls its extracted insights and past experiences to make informed decisions. Our empirical results highlight the robust learning efficacy of the ExpeL agent, indicating a consistent enhancement in its performance as it accumulates experiences. We further explore the emerging capabilities and transfer learning potential of the ExpeL agent through qualitative observations and additional experiments.
Overview
AI snapshot
Written from the paper’s full text.
TL;DR
ExpeL is an LLM agent that learns from its own task experiences without parameter updates by storing trajectories, extracting natural-language insights, and retrieving similar successful trajectories for inference. It improves over ReAct/Act baselines across HotpotQA, ALFWorld, and WebShop and shows forward transfer from HotpotQA to FEVER.
Key findings
- ExpeL consistently outperformed ReAct and Act across HotpotQA, ALFWorld, and WebShop; in ablations, ExpeL reached 39.0 ± 1.7 success on HotpotQA vs. ReAct 28.0 ± 1.4, and 59.0 ± 0.3 on ALFWorld vs. ReAct 40.0 ± 0.3.
- Both insight extraction and retrieval mattered: insights-only/retrieval-only variants scored 36%/31% on HotpotQA, 50%/55% on ALFWorld, and 37%/38% on WebShop, with WebShop scores of 0.675/0.67, respectively.
- Cross-task learning matched Reflexion on HotpotQA (40% at R3 vs. 39%) and outperformed it on ALFWorld (54% at R3 vs. 59%) without repeated attempts.
- Transfer from HotpotQA to FEVER improved performance for both transfer agents, with the agent using few in-context examples improving more; ablations showed learned insights beat hand-crafted ones (32.0 ± 1.1), reflections hurt (29.0 ± 0.4), gpt-3.5 insights were lower (32.0 ± 0.4), and task-similarity retrieval was best (59.0 ± 0.3 vs. reasoning similarity 48.5 ± 2.1 and random 42.5 ± 0.8).
Methods
- ExpeL uses ReAct as its base planner; during training it gathers experiences by retrying tasks up to Z times with Reflexion, stores trajectories in an experience pool, and extracts natural-language insights via LLM operators ADD/EDIT/UPVOTE/DOWNVOTE from fail/success pairs and lists of successes. At evaluation, it augments task specifications with insights and retrieves top-k successful trajectories by task similarity using Faiss/kNN/all-mpnet-base-v2.
- Evaluated on HotpotQA, ALFWorld, WebShop, and FEVER transfer; four-fold validation, success rate (exact match for HotpotQA/FEVER, task completion for ALFWorld, matching item for WebShop), plus WebShop reward and ALFWorld task-type breakdown. Baselines were ReAct, Act, and imitation learning; actions used gpt-3.5-turbo-0613, insights used gpt-4-0613, with temperature 0 and greedy decoding.
- Transfer used HotpotQA as source and FEVER as target, adapting source insights with target fewshot examples and comparing against ReAct, Act, and transfer without task demonstrations.
Limitations
- Stated: the work investigates textual observations only, limiting real-world applicability; closed-source API LLMs may be off-limits in some applications; insights currently fit the context window but lifelong learning may need extra retrieval; prompting techniques lack the theoretical underpinnings of reinforcement learning.
- Evident: only deterministic environments are considered; evaluation is limited to four text-based benchmarks with four-fold validation; transfer assumes source and target tasks share common knowledge.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
28 Own interpretation · 27 Own result · 9 No basis given · 6 Partly supported · 3 Supported · 3 Sources not found · 2 Not checked yet · accuracy not yet measured
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“ExpeL matches Reflexion's performance (40% at R3 vs. 39%) for HotpotQA and even outperforms it for ALFWorld (54% at R3 vs. 59%) without repeated attempts.”
§ ExpeL: LLM Agents Are Experiential Learners > 5.2 Main Results
What the sources say
- Partly supportedcites Shinn et al. 2023 sourceExperimental — accuracy not yet measured
Section 4.2 and Figure 4 describe Reflexion’s HotPotQA evaluation, so the cited paper supports the identification of Reflexion as the comparator. But our copy does not report 39% as Reflexion’s R3 performance or compare it with ExpeL’s 40%. The Results section’s 39% instead refers to questions the CoT (GT) agent could not answer. (LLM, conf=0.84)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“ExpeL matches Reflexion's performance (40% at R3 vs. 39%) for HotpotQA and even outperforms it for ALFWorld (54% at R3 vs. 59%) without repeated attempts.”
§ ExpeL: LLM Agents Are Experiential Learners > 5.2 Main Results
What the sources say
- Partly supportedcites Shinn et al. 2023 sourceExperimental — accuracy not yet measured
Section 4.1 and the ALFWorld Results establish that Reflexion was evaluated on ALFWorld and improved across repeated trials. Our copy does not give the cited 54% at R3 figure or compare Reflexion with ExpeL, so it cannot substantiate the claimed performance comparison or ExpeL’s lack of repeated attempts. (LLM, conf=0.83)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“We evaluated ExpeL on three vastly different domains and consistently outperformed strong baselines.”
- Also stated in: § 5.2 Main Results · plain
“Experiential learning Augmenting agents with abstracted insights and the ability to recall successful trajectories improve performance across all environments compared to baseline agents.”
§ ExpeL: LLM Agents Are Experiential Learners > Tom Mitchell
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Also stated in: § 5.2 Main Results · plain
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Our empirical results highlight the robust learning efficacy of the ExpeL agent, indicating a consistent enhancement in its performance as it accumulates experiences.”
§ ExpeL: LLM Agents Are Experiential Learners > Abstract
hedgeddirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“We demonstrated its learning abilities by showing its performance gain compared to vanilla ReAct and Act agents.”
§ ExpeL: LLM Agents Are Experiential Learners > 6 Conclusion and Limitations
boosteddirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“We noticed our ExpeL agent updated its beliefs through the insights and over its gained experience.”
§ ExpeL: LLM Agents Are Experiential Learners > Hypothesis Formulation & Constraints Adaptation
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Results in Tab. 3 show several significant findings: (1) learned insights by the agent are more advantageous than hand-crafted ones;”
§ ExpeL: LLM Agents Are Experiential Learners > The Thirty-Eighth AAAI Conference on Artificial Intelligence (AAAI-24)
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“For example, in ALFWorld, the agent completely changed the priors it had in ReAct on the likely locations of a pan (from drawers/countertops/cabinets to stoveburners).”
§ ExpeL: LLM Agents Are Experiential Learners > Hypothesis Formulation & Constraints Adaptation
plaindirectSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“Furthermore, we investigated a transfer learning scenario where extracting insights from a set of source tasks can benefit the ExpeL agent in solving a target task.”
§ ExpeL: LLM Agents Are Experiential Learners > 6 Conclusion and Limitations
plainconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Although ReAct was sometimes not able to reassess its situation when attempting to solve a task, Ex-peL demonstrated its proficiency in identifying and rectifying missteps.”
§ ExpeL: LLM Agents Are Experiential Learners > Hypothesis Formulation & Constraints Adaptation
boostedSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Probably not a substantial claim. 18
- It may state two separate claims as one. 24
- The quote only partly states this claim. 1
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Retrieval allows LLMs to access databases, mitigating hallucinations (Li et al. 2022;Wang, Yang, and Wei 2023;Rubin, Herzig, and Berant 2022;Liu et al. 2022).”
§ ExpeL: LLM Agents Are Experiential Learners > Tom Mitchell
What the sources say
- Not matchedcites Li et al. 2022
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites Wang, Yang, and Wei 2023
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Partly supportedcites Rubin, Herzig, and Berant 2022 sourceExperimental — accuracy not yet measured
The Background: Prompt Retrieval Problem setup section describes retrieving examples from a training set and supplying them to a language model as a prompt. This supports retrieval of stored examples, but the paper does not establish that retrieval gives LLMs access to databases or mitigates hallucinations. (LLM, conf=0.81)
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
- Not matched
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“A relatively less explored area is facilitating agents to learn autonomously from their own experiences, similar to a student gaining insights from practicing for an exam.”
§ ExpeL: LLM Agents Are Experiential Learners > 4 ExpeL: An Experiential Learning Agent
What the sources say
- Not matchedcites Lin et al. 2023a
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Partly supportedcites Shaw et al. 2023 sourceExperimental — accuracy not yet measured
Section 3.1 (page not available) describes improving an agent iteratively using successful episodes from its own tree-search policy, supporting the idea of learning from experience. It does not say that autonomous learning from experience is a relatively less explored area; it notes that prior work has explored reinforcement learning for agent improvement. (LLM, conf=0.79)
- Supportedcites Shinn et al. 2023 source
The Introduction says earlier LLM-agent approaches relied on in-context examples to teach agents. The Related work and Self-reflection sections present Reflexion as a way for agents to identify their own errors, retain those experiences, and improve across trials. Together, these support the claim that autonomous learning from experience was a relatively less explored direction. (LLM, conf=0.88)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
- Not matched
Every source the paper cites for this claim was checked and supports it.
In the paper
“While Reflexion improves results by iteratively refining insights through repeated task execution (R1, R2, R3...), our ExpeL agent leverages cross-task learning by accumulating task experience.”
§ ExpeL: LLM Agents Are Experiential Learners > 5.2 Main Results
What the sources say
- Supportedcites Shinn et al. 2023 source
The Introduction describes Reflexion using feedback from one episode to improve the next attempt. The Self-reflection section describes an iterative loop in which the agent stores a verbal summary after each trial, and the Results report improvement over consecutive trials. The sentence's description of ExpeL is the citing authors' own comparison, not a claim attributed to this source. (LLM, conf=0.93)
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“Furthermore, self-improvement methods like Reflexion (Shinn et al. 2023) facilitate intra-task improvements, but ExpeL enables inter-task learning.”
§ ExpeL: LLM Agents Are Experiential Learners > 4.5 ExpeL's Strengths
What the sources say
- Supportedcites Shinn et al. 2023 source
The Self-reflection section describes Reflexion as a loop of trials for a task, using feedback from each attempt to improve the next. This supports the cited characterization of Reflexion as facilitating intra-task improvement; the separate claim about ExpeL is not attributed to this source. (LLM, conf=0.92)
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Agents like Reflexion showcase feedback-based improvement, yet often lack cross-task memory (Shinn et al. 2023).”
§ ExpeL: LLM Agents Are Experiential Learners > Tom Mitchell
What the sources say
- Partly supportedcites Shinn et al. 2023 sourceExperimental — accuracy not yet measured
The Introduction and Self-reflection sections describe Reflexion improving through feedback and storing reflections for later trials. They describe memory across trials, but do not establish that Reflexion lacks cross-task memory; that limitation is unsupported by our copy. (LLM, conf=0.83)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Partly supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Additionally, previous works on learning in LLM agents have primarily been trained on extensive human-labeled datasets (Lin et al. 2023a;Shaw et al. 2023) or improved via iterative retries (Shinn et al. 2023) on a single task.”
§ ExpeL: LLM Agents Are Experiential Learners > 4 ExpeL: An Experiential Learning Agent
What the sources say
- Not matchedcites Lin et al. 2023a
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Supportedcites Shaw et al. 2023 source
The Human Demonstrations section reports that the authors started with approximately 1.3 million human demonstrations for MiniWoB++ tasks, and §3.1 says they trained their GUI agent on human demonstrations. Shaw et al. is cited for the human-labeled-data part of the sentence; the iterative-retries part is attributed to Shinn et al. (LLM, conf=0.90)
- Supportedcites Shinn et al. 2023 source
The Results section says Reflexion agents gather experience and retry failed tasks until three consecutive failures on the particular task. The Introduction explains that feedback from one attempt guides the agent in the next, supporting the description of iterative retries on a task. The human-labeled-dataset part of the sentence is attributed to other citations, not Shinn et al. (LLM, conf=0.92)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Not matched
Every source the paper cites for this claim was checked and supports it.
In the paper
“However, since current LLMs are bounded by context window size (Tworkowski et al. 2023), these agents have no recollections of what they have seen, and therefore no learning can be done outside of a few demonstrations.”
§ ExpeL: LLM Agents Are Experiential Learners > Tom Mitchell
What the sources say
- Supportedcites Tworkowski et al. 2023 source
The citation is attached to the claim that LLMs have bounded context windows. Section 4.4 describes a standard 2K-token context length that limits how many demonstrations fit, and Section 4 describes extending existing models beyond their original context length. The statements about agents’ recollections and learning are the citing authors’ inference, not the part attributed to this source. (LLM, conf=0.91)
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Supported
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“Furthermore, since our extracted insights do not exceed the current LLM's token limit, we can fit them into the agent's context window.”
§ ExpeL: LLM Agents Are Experiential Learners > 6 Conclusion and Limitations
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds. It may state two separate claims as one.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“Lastly, we presented several unexpected emerged abilities our agent developed at the end of its training.”
§ ExpeL: LLM Agents Are Experiential Learners > 6 Conclusion and Limitations
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“This ability was particularly observed in HotpotQA where a likely influential insight was stating that the agent should “consider the answer might be in the observations already made”.”
§ ExpeL: LLM Agents Are Experiential Learners > Hypothesis Formulation & Constraints Adaptation
hedgedconditionalSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“This behavior emerged from the extracted insight claiming that “when searching for an item” it needs to “consider its nature and its typical usage”, leading the agent to promptly and accurately find the correct item at the first step while the ReAct agent could not find it in time.”
§ ExpeL: LLM Agents Are Experiential Learners > Hypothesis Formulation & Constraints Adaptation
plaindirectSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The results demonstrate a notable improvement in the success rate when ExpeL is paired with Reflexion, with the success rate increasing as the number of task reattempts grows.”
§ ExpeL: LLM Agents Are Experiential Learners > 5.5 ExpeL with Task Reattempts
boosteddirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Reason similarity is still advantageous but slightly drops in performance, possibly due to dynamically changing fewshots during a single trajectory, causing instabilities.”
§ ExpeL: LLM Agents Are Experiential Learners > The Thirty-Eighth AAAI Conference on Artificial Intelligence (AAAI-24)
hedgeddirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“We empirically found that gpt-4-0613 is better than gpt-3.5-turbo-0613at following instructions on how to use the insight extraction operators and hallucinated less.”
- Also stated in: § 4.5 ExpeL's Strengths · boosted
“our experiments show that using gpt-4 to extract insights outperforms gpt-3.5-turbo(refer to Sec. 5.6).”
§ ExpeL: LLM Agents Are Experiential Learners > 4.2 Learning from Experiences
boosteddirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The quote only partly states this claim.
- Also stated in: § 4.5 ExpeL's Strengths · boosted
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Randomly selected baseline has a significant drop in performance while ranking using reason similarity also has a noticeable dip.”
§ ExpeL: LLM Agents Are Experiential Learners > The Thirty-Eighth AAAI Conference on Artificial Intelligence (AAAI-24)
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Hand-crafted insights enjoyed a performance boost over ReAct but were less effective than LLMgenerated ones.”
§ ExpeL: LLM Agents Are Experiential Learners > The Thirty-Eighth AAAI Conference on Artificial Intelligence (AAAI-24)
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“This highlights ExpeL's capacity to recover from errors and stay on course without hallucinating when completing tasks.”
§ ExpeL: LLM Agents Are Experiential Learners > Hypothesis Formulation & Constraints Adaptation
plainSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
In the paper
“Rather than altering the LLM parameters, adjusting the prompts may be more beneficial: this strategy ensures that the LLM's inherent common sense knowledge remains intact, allowing for superior generalization (Liu et al. 2023a).”
§ ExpeL: LLM Agents Are Experiential Learners > 4 ExpeL: An Experiential Learning Agent
What the sources say
- Not matchedcites Liu et al. 2023a
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
hedgeddirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Not matched
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“It is not restricted to specific language models and complements existing strategies aimed at enhancing LLM agent planning capabilities.”
§ ExpeL: LLM Agents Are Experiential Learners > 4.5 ExpeL's Strengths
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds. It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“ExpeL does not rely on retries during deployment, which certain domains require.”
§ ExpeL: LLM Agents Are Experiential Learners > 4.5 ExpeL's Strengths
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“To highlight, this trial and error way of gathering experiences not only improves the chances of getting more positive examples for experience recall during evaluation but also allows for collecting valuable success/failure pairs used for comparisons during insight extraction (Sec.4.2).”
§ ExpeL: LLM Agents Are Experiential Learners > 4.1 Gathering Experiences
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: unclear — It is unclear why the paper says it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“This behavior is possibly encouraged by the generated insight “reassess the situation and consider alternative actions” if “an attempt does not progress the task”.”
§ ExpeL: LLM Agents Are Experiential Learners > Hypothesis Formulation & Constraints Adaptation
hedgedconditionalSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds. It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“First, we let the agent compare a failed trajectory with a successful trajectory for the same task.This comparison offers a concrete understanding of the agent's shortcomings, highlighting the correct and incorrect actions.”
§ ExpeL: LLM Agents Are Experiential Learners > 4.2 Learning from Experiences
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“(2) using reflections in addition to success/failure pairs and lists of successes is disadvantageous, possibly due to reflections sometimes outputting hallucinations, therefore misleading the insight extraction stage;”
§ ExpeL: LLM Agents Are Experiential Learners > The Thirty-Eighth AAAI Conference on Artificial Intelligence (AAAI-24)
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“However, extra retrieval steps for insights might be needed for truly lifelong learning agents to ensure a manageable context window size.”
§ ExpeL: LLM Agents Are Experiential Learners > 6 Conclusion and Limitations
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence. The paper’s conclusions may not depend on it.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Additionally, our learning approach is highly accessible; it demands less data, reduces computational resources, and is straightforward to implement.”
§ ExpeL: LLM Agents Are Experiential Learners > 4.5 ExpeL's Strengths
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds. It may state two separate claims as one.
In the paper
“Furthermore, finetuning an LLM restricts its functionalities and can hurt its generalization abilities (Du et al. 2022).”
§ ExpeL: LLM Agents Are Experiential Learners > Tom Mitchell
What the sources say
- Not matchedcites Du et al. 2022
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plainconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
- Not matched
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“In this work, we investigated tasks with textual observation, which is limiting in real-world scenarios.”
§ ExpeL: LLM Agents Are Experiential Learners > 6 Conclusion and Limitations
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Reason similarity is still advantageous but slightly drops in performance, possibly due to dynamically changing fewshots during a single trajectory, causing instabilities.”
§ ExpeL: LLM Agents Are Experiential Learners > The Thirty-Eighth AAAI Conference on Artificial Intelligence (AAAI-24)
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“First and foremost, ExpeL offers inherent interpretability, as both the extracted experiences and successful trajectories are presented in natural language.This design allows users to easily inspect, modify, or remove potentially harmful trajectories/insights -a challenge in finetuned models.”
§ ExpeL: LLM Agents Are Experiential Learners > 4.5 ExpeL's Strengths
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Second, we let the agent identify patterns within a set of successful trajectories from different tasks.This approach sheds light on common”
§ ExpeL: LLM Agents Are Experiential Learners > 4.2 Learning from Experiences
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Lastly, unlike reinforcement learning methods, prompting techniques lack theoretical underpinnings that could potentially impact the efficiency of the resulting policies.”
§ ExpeL: LLM Agents Are Experiential Learners > 6 Conclusion and Limitations
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: fails — The paper may give no reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“This belief thereby update enables the agent to avoid unnecessary actions and increase efficiency in solving a given task.”
§ ExpeL: LLM Agents Are Experiential Learners > Hypothesis Formulation & Constraints Adaptation
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“These observations highlight the synergistic interplay between abstraction and recollection in experiential learning, with ExpeL showing a quantitative advantage over baseline/restricted learning mode agents.”
§ ExpeL: LLM Agents Are Experiential Learners > 5.2 Main Results
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Additionally, we investigated the efficacy of our method by using closed-source API LLMs, which can be off-limits in some applications.”
§ ExpeL: LLM Agents Are Experiential Learners > 6 Conclusion and Limitations
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Additionally, we compare ExpeL with ExpeL (retrieve-only) and ExpeL (insights-only) to highlight that both insight extraction and task similarity retrieval are essential and synergistic.”
§ ExpeL: LLM Agents Are Experiential Learners > 6 Conclusion and Limitations
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
The authors state this as an assumption.
The source and target tasks exhibit common knowledge.
“Like most transfer learning settings, we assume that the source and target tasks exhibit common knowledge.”
The paper relies on this without support.
Some of the most potent language models are proprietary.
“Furthermore, some of the most potent language models are proprietary.”
The paper relies on this without support.
WebShop presents a unique challenge, requiring both website-based reasoning and precise execution of actions.
“Furthermore, WebShop presents a unique challenge, requiring both website-based reasoning (price comparisons, query reformulation, etc.) and precise execution of actions (searching, clicking, option selection, etc.).”
The paper relies on this without support.
Human learning occurs mainly either by storing successful trajectories in memory or by extracting high-level insights from experiences.
“Human learning occurs mainly either by storing successful trajectories in memory, which can be later recalled as specific examples, or by extracting high-level insights from experiences, enabling generalization to novel situations.”
The paper relies on this without support.
The source and target tasks are inherently different, so no experience pool can be retrieved from for transfer.
“Notice that since source and target tasks are inherently different, we do not have an experience pool to retrieve from; thus, the ExpeL Transfer agents use the existing fixed fewshot examples as in-context examples.”
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“In contrast, agent memory systems are instantiated within an agent's ongoing interaction with an environment , continuously incorporating new information generated by the agent's own actions and environmental feedback into a persistent memory base (Wang et al., 2024m; Zhao et al., 2024; Sun et al., 2025e).”
From Memory in the Age of AI Agents (2025) · cites this paper as Zhao et al., 2024
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“The resulting summaries capture the global, high-level information of the original data, rather than specific factual or experiential details (Zhao et al., 2024; Anokhin et al., 2024).”
From Memory in the Age of AI Agents (2025) · cites this paper as Zhao et al., 2024
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“In contrast, agent memory systems are instantiated within an agent's ongoing interaction with an environment , continuously incorporating new information generated by the agent's own actions and environmental feedback into a persistent memory base (Wang et al., 2024m; Zhao et al., 2024; Sun et al., 2025e).”
From Memory in the Age of AI Agents (2025) · cites this paper as Zhao et al., 2024
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“The resulting summaries capture the global, high-level information of the original data, rather than specific factual or experiential details (Zhao et al., 2024; Anokhin et al., 2024).”
From Memory in the Age of AI Agents (2025) · cites this paper as Zhao et al., 2024
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Most existing approaches construct an experience pool or memory bank and focus on improving the retrieval of useful past information (Zhao et al., 2024; Gao et al., 2024; Xu et al., 2025b).”
From 3DLLM-Mem: Long-Term Spatial-Temporal Memory for Embodied 3D Large Language Model (2025) · cites this paper as Zhao et al., 2024
Experimental — accuracy not yet measured
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Recent agentic-memory systems pursue a similar idea for LLM agents. After solving a problem, an agent can distill its trajectory into a textual abstraction of the skill it has learned, store this abstraction in persistent memory, and refine it as more related problems are encountered (Shinn et al., 2023; Zhao et al., 2024; Zhang et al., 2025; Sumers et al., 2024; Park et al., 2023; Majumder et al.”
From Useful Memories Become Faulty When Continuously Updated by LLMs (2026) · cites this paper as Zhao et al., 2024
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“The dominant approach is to curate memory at write time.”
From Just-in-Time Memory: Learning to Curate Task-Adaptive Memory for LLM Agents (2026) · cites this paper as Zhao et al., 2024
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“This demands a curated set of training tasks drawn from the target task distribution, often paired with a verifier that scores each attempt.”
From DAEDALUS: Bootstrapping Agent Memory from Self-Generated Tasks (2026) · cites this paper as Zhao et al., 2024
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“The other five build memory from the benchmark training tasks, and therefore have access to information that DAEDALUS does not.”
From DAEDALUS: Bootstrapping Agent Memory from Self-Generated Tasks (2026) · cites this paper as Zhao et al., 2024
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Existing methods typically extract knowledge from accumulated trajectories via reflection, memory, rules, or skills.”
From Evo-Harness: Context-to-Harness Skill Compilation for Self-Evolving Agents (2026) · cites this paper as Zhao et al., 2024
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“ExpeL extracts and retrieves insights from previous tasks (Zhao et al., 2024)”
From GitSwarm: Decentralized Compounding Inference (2026) · cites this paper as Zhao et al., 2024
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“In contrast, agent memory defines a substantially broader cognitive scope. Beyond transient context assembly, it encompasses the persistent storage of factual knowledge (Zhong et al., 2024), the accumulation and evolution of experiential traces (Zhao et al., 2024; Tang et al., 2025d; Zhang et al., 2025d), and, in some cases, the internalization of memory into model parameters (Wang et al., 2025o).”
From Memory in the Age of AI Agents (2025) · cites this paper as Zhao et al., 2024
The cited source says what the claim attributes to it.
In the citing paper
“Consequently, such adaptation minimizes redundant computations and refines decision-making over time (Zhao et al., 2024; Shinn et al., 2023b).”
From Memory in the Age of AI Agents (2025) · cites this paper as Zhao et al., 2024
The cited source says what the claim attributes to it.
In the citing paper
“ExpeL (Zhao et al., 2024) autonomously gathers experience through trial-and-error, storing successful trajectories as exemplars while extracting textual insights to guide future actions.”
From Memory in the Age of AI Agents (2025) · cites this paper as Zhao et al., 2024
The cited source says what the claim attributes to it.
In the citing paper
“For example, ExpeL (Zhao et al., 2024) synergizes concrete trajectories with abstract textual insights, allowing agents to recall specific solutions while applying general heuristics.”
From Memory in the Age of AI Agents (2025) · cites this paper as Zhao et al., 2024
The cited source says what the claim attributes to it.
In the citing paper
“Combining both, ExpeL (Zhao et al., 2024), From Experience to Strategy (Xia et al., 2025), and ReMe (Cao et al., 2025b) contrast successful and failed experiences to uncover holistic planning insights.”
From Memory in the Age of AI Agents (2025) · cites this paper as Zhao et al., 2024
The cited source says what the claim attributes to it.
In the citing paper
“In contrast, agent memory defines a substantially broader cognitive scope. Beyond transient context assembly, it encompasses the persistent storage of factual knowledge (Zhong et al., 2024), the accumulation and evolution of experiential traces (Zhao et al., 2024; Tang et al., 2025d; Zhang et al., 2025d), and, in some cases, the internalization of memory into model parameters (Wang et al., 2025o).”
From Memory in the Age of AI Agents (2025) · cites this paper as Zhao et al., 2024
The cited source says what the claim attributes to it.
In the citing paper
“ExpeL (Zhao et al., 2024) and AutoAgent (Wang et al., 2026) operate at two levels simultaneously, but both use predetermined levels without adaptive selection: two-speed systems, not continuous spectra.”
From Experience Compression Spectrum: Unifying Memory, Skills, and Rules in LLM Agents (2026) · cites this paper as Zhao et al., 2024
The cited source says what the claim attributes to it.
In the citing paper
“Recent works show that memorizing task procedures, interface affordances, and hidden failure modes improve agent performance at inference time [Wang et al., 2024a, Zhao et al., 2024, Bouzenia et al., 2025, Wang et al., 2025b, Tang et al., 2025].”
From LongMemEval-V2: Evaluating Long-Term Agent Memory Toward Experienced Colleagues (2026) · cites this paper as Zhao et al., 2024
The cited source says what the claim attributes to it.
In the citing paper
“Systematically extracts success/failure 'rules of thumb' from trajectory comparisons.”
From Memory for Autonomous LLM Agents:Mechanisms, Evaluation, and Emerging Frontiers (2026) · cites this paper as Zhao et al., 2024
The cited source says what the claim attributes to it.
In the citing paper
“These methods retain an insight based on the outcome of the trajectory it came from, without confirming that it helps the agent overcome the failure; DAEDALUS instead keeps an insight only once it has turned a failure into repeated success.”
From DAEDALUS: Bootstrapping Agent Memory from Self-Generated Tasks (2026) · cites this paper as Zhao et al., 2024
The cited source says what the claim attributes to it.
In the citing paper
“ExpeL (Zhao et al., 2024) and AutoGuide (Fu et al., 2024) contrast successful and failed retries of each training task to extract global insights and contextspecific guidelines, respectively.”
From DAEDALUS: Bootstrapping Agent Memory from Self-Generated Tasks (2026) · cites this paper as Zhao et al., 2024
The cited source says what the claim attributes to it.
In the citing paper
“Reflection and experiential-learning methods retain lessons or heuristics derived from previous trajectories, while systems such as ACE and SkillOpt maintain editable playbooks or skill documents (Shinn et al., 2023;Zhao et al., 2024;Zhang et al., 2026c;Yang et al., 2026).”
From ADIAS: Automated Design of Interactive Agentic Systems (2026) · cites this paper as Zhao et al., 2024
Citations
Cites
Language Models are Few-Shot Learners
Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al. (2020) · arXiv · doi:10.48550/arxiv.2005.14165 · Cited by 23,372
(2011) · Cited by 14,317
Watkins, C. J. C. H., & Dayan, P. (1992) · Machine Learning · doi:10.1007/bf00992698 · Cited by 9,083
A Comprehensive Survey on Transfer Learning
Zhuang, F., Qi, Z., Duan, K., Xi, D., Zhu, Y., Zhu, H., Xiong, H., & He, Q. (2021) · Proceedings of the IEEE · doi:10.1109/jproc.2020.3004555 · Cited by 6,678
LLaMA: Open and Efficient Foundation Language Models
Touvron, H., Lavril, T., Izacard, G., Martinet, X., Lachaux, M.-A., Lacroix, T., Rozière, B., Goyal, N., Hambro, E., Azhar, F., et al. (2023) · arXiv · doi:10.48550/arxiv.2302.13971 · Cited by 4,602
Chain-Of-Thought Prompting Elicits Reasoning in Large Language Models
Wei, J., Wang, X., Schuurmans, D., Bosma, M., Ichter, B., Xia, F., Chi, E., Le, Q. V., & Zhou, D. (2022) · Advances in Neural Information Processing Systems 35 · doi:10.52202/068431-1800 · Cited by 4,505
Liu, P., Yuan, W., Fu, J., Jiang, Z., Hayashi, H., & Neubig, G. (2023) · ACM Computing Surveys · doi:10.1145/3560815 · Cited by 4,361
Liu, P., Yuan, W., Fu, J., Jiang, Z., Hayashi, H., & Neubig, G. (2021) · arXiv · doi:10.48550/arxiv.2107.13586 · Cited by 4,352
Billion-scale similarity search with GPUs
Johnson, J., Douze, M., & Jegou, H. (2021) · IEEE Transactions on Big Data · doi:10.1109/tbdata.2019.2921572 · Cited by 4,228
PaLM: Scaling Language Modeling with Pathways
Chowdhery, A., Narang, S., Devlin, J., Bosma, M., Mishra, G., Roberts, A., Barham, P., Chung, H. W., Sutton, C., Gehrmann, S., et al. (2022) · arXiv · doi:10.48550/arxiv.2204.02311 · Cited by 2,590
Research works printed but not linked: 5
- NeurIPS.
- Du, M.; He, F.; Zou, N.; Tao, D.; and Hu, X. 2022. Shortcut Learning of Large Language Models in Natural Language Understanding: A Survey. arXiv preprint.
- NeurIPS.
- Maas; Carey; Wheeler; Saatchi; Billington; and Shamash. 2023. To Infinity and Beyond: SHOW-1 and Showrunner Agents in Multi-Agent Simulations. arXiv preprint.
- Touvron, H.; Martin, L.; Stone, K.; Albert, P.; Almahairi, A.; Babaei, Y.; Bashlykov, N.; Batra, S.; Bhargava, P.; Bhosale, S.; et al. 2023b. Llama 2: Open Foundation and Fine-Tuned Chat Models. arXiv preprint.
Cited by
Sorted by citations
The Rise and Potential of Large Language Model Based Agents: A Survey
Xi, Z., Chen, W., Guo, X., He, W., Ding, Y., Hong, B., Zhang, M., Wang, J., Jin, S., Zhou, E., et al. (2023) · arXiv · doi:10.48550/arxiv.2309.07864 · Cited by 972
The rise and potential of large language model based agents: a survey
Xi, Z., Chen, W., Guo, X., He, W., Ding, Y., Hong, B., Zhang, M., Wang, J., Jin, S., Zhou, E., et al. (2025) · Science China Information Sciences · doi:10.1007/s11432-024-4222-0 · Cited by 970
DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelines
Khattab, O., Singhvi, A., Maheshwari, P., Zhang, Z., Santhanam, K., Vardhamanan, S., Haq, S., Sharma, A., Joshi, T. T., Moazam, H., et al. (2023) · arXiv · doi:10.48550/arxiv.2310.03714 · Cited by 911
A Survey on the Memory Mechanism of Large Language Model-based Agents
Zhang, Z., Dai, Q., Bo, X., Ma, C., Li, R., Chen, X., Zhu, J., Dong, Z., & Wen, J.-R. (2025) · ACM Transactions on Information Systems · doi:10.1145/3748302 · Cited by 592
ChatDev: Communicative Agents for Software Development
Qian, C., Liu, W., Liu, H., Chen, N., Dang, Y., Li, J., Yang, C., Chen, W., Su, Y., Cong, X., et al. (2024) · Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2024.acl-long.810 · Cited by 490
Metadata
- Authors
- Zhao, A., Huang, D., Xu, Q., Lin, M., Liu, Y.-J., & Huang, G.
- Affiliations
- Tsinghua University (CN)
- Venue
- Proceedings of the AAAI Conference on Artificial Intelligence
- Published
- 2024-03-24
- Volume
- 38
- Issue
- 17
- Pages
- 19632-19642
- ISSN
- 2159-5399 (ISSN-L)
- 2374-3468
- OpenAlex
- W4393160747
- Cited by
- 153
- References
- 31
- Citations by year
- 2026: 61 · 2025: 63 · 2024: 28
- FWCI
- 44.76
- Type
- conference paper
- Language
- en
- Open access
- diamond
- doi.org
- Locations
- Proceedings of the AAAI Conference on Artificial Intelligence (publishedVersion)
- arXiv (Cornell University) (submittedVersion, cc-by)
- AAAI Publications (The Association for the Advancement of Artificial Intelligence (AAAI)) (submittedVersion)
- arXiv (Cornell University) (cc-by)
- Funders
- National Natural Science Foundation of China (U2336214, 62332019, 62022048)
- Tsinghua University
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › AI-based Problem Solving and Planning
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Keywords
- large language models, LLM agents, experiential learning, decision-making, transfer learning, nonparametric learning
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
A language model drew the claims from the paper on 2026-10-16, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-16 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
