The DOI matches its Crossref record; 84 of the 85 references it cites were found in a registry or an index; 6 of the 20 sources its claims cite were checked for support.
MedAgents: Large Language Models as Collaborators for Zero-shot Medical Reasoning
Open accessAbstract
Large language models (LLMs), despite their remarkable progress across various general domains, encounter significant barriers in medicine and healthcare.This field faces unique challenges such as domain-specific terminologies and reasoning over specialized knowledge.To address these issues, we propose MEDAGENTS, a novel multi-disciplinary collaboration framework for the medical domain.MedAgents leverages LLM-based agents in a role-playing setting that participate in a collaborative multi-round discussion, thereby enhancing LLM proficiency and reasoning capabilities.This training-free framework encompasses five critical steps: gathering domain experts, proposing individual analyses, summarising these analyses into a report, iterating over discussions until a consensus is reached, and ultimately making a decision.Our work focuses on the zeroshot setting, which is applicable in realworld scenarios.Experimental results on nine datasets (MedQA, MedMCQA, PubMedQA, and six subtasks from MMLU) establish that our proposed MEDAGENTS framework excels at mining and harnessing the medical expertise within LLMs, as well as extending its reasoning abilities.Our code can be found at https: //github.com/gersteinlab/MedAgents.① Expert Gathering ② Analysis Proposition ③ Report Summarization ④ Collaborative Consultation ⑤ Decision Making A 66-year-old male with a history of heart attack and recurrent stomach ulcers is experiencing persistent cough and chest pain, and recent CT scans indicate a possible lung tumor.Designing a treatment plan that minimizes risk and maximizes outcomes is the current concern due to his deteriorating health and medical history....... ...... .
Overview
AI snapshot
Written from the paper’s full text.
TL;DR
MedAgents is a training-free, multi-agent role-playing framework for zero-shot medical question answering that simulates multi-disciplinary expert consultation. Across nine medical datasets, it outperforms zero-shot baselines and approaches few-shot chain-of-thought/self-consistency baselines.
Key findings
- MedAgents achieved 64.1% on MedQA and 59.3% on MedMCQA with GPT-3.5, and 83.7% on MedQA and 74.8% on MedMCQA with GPT-4, outperforming zero-shot baselines and matching or approaching strong few-shot CoT+SC baselines.
- Ablation on MedQA showed gains from each stage: direct prompting 49.0%, CoT prompting 55.0%, +Analysis 62.0%, +Analysis & Summarization 65.0%, +Analysis & Summarization & Consultation 67.0%.
- Human error analysis of 40 randomly selected error cases in MedQA and MedMCQA found 77% of errors involved confusion about domain knowledge, while 8% were due to CoT errors.
- Agent quantity/domain variation mattered: on GPT-3.5, MedAgents with 6 different domains scored 64.1/59.3 on MedQA/MedMCQA, versus 59.2/58.1 with 6 same domains, 57.5/57.3 with 5 same domains, and 55.9/57.0 with 4 same domains.
Methods
- The framework has five stages: expert gathering, analysis proposition, report summarization, collaborative consultation, and decision making; experts vote and revise a report iteratively until consensus or a maximum attempt threshold.
- Experiments used GPT-3.5-Turbo and GPT-4 in zero-shot settings on nine datasets: MedQA, MedMCQA, PubMedQA, and six medical MMLU subtasks; 300 examples were randomly sampled per dataset.
- Generation settings included temperature 1.0 and top_p 1.0; question/option expert numbers were m=5, n=2 except PubMedQA m=4, n=2; maximum consultation attempts t=5; baselines included zero-shot, few-shot, CoT, and self-consistency prompting.
Limitations
- The paper states that parameterized LLM knowledge may need updating over time, that integrating diverse models at different stages could be explored, and that the framework may have limited applicability in low-resource languages.
- The paper reports cost and latency: about $1.41 per 100 QA examples and about 40 seconds inference time per example, which the authors acknowledge is more costly than CoT or direct prompting.
- The text makes evident a domain-knowledge bottleneck: 77% of analyzed errors related to lack or mis-retrieval of domain knowledge, and the domain variation study was conducted on only 20 samples due to manual evaluation.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
19 Own result · 16 Own interpretation · 4 Partly supported · 3 Sources not found · 2 No basis given · 1 No full text of the sources · 1 Not checked yet · 1 Own method · accuracy not yet measured
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“We have found that reliance on CoT in isolation can inadvertently result in hallucinations -spurious outputs typically associated with the misapplication of medical terminologies.”
- Also stated in: § A Analysis on the Addition of CoT · hedged
“As demonstrated in the example, for specialist domains that demand considerable expert knowledge such as the medical domain knowledge, employing a CoT approach might sometimes lead to hallucination (Bubeck et al., 2023;Guerreiro et al., 2023;Ji et al., 2023;Maynez et al., 2020).”
§ MEDAGENTS: Large Language Models as Collaborators for Zero-shot Medical Reasoning > 3.2 Main Results
What the sources say
- Partly supportedcites Bubeck et al., 2023 sourceExperimental — accuracy not yet measured
Section 9.1 describes LLM hallucinations, and Section 10.2 warns that relying on hallucinated factual claims can be costly in healthcare. Figure 1.8 also shows a medical-note example containing claims not established by the patient's facts. The cited passages do not link those errors to reliance on CoT in isolation or characterize them as misapplied medical terminology. (LLM, conf=0.85)
- Partly supportedcites Guerreiro et al., 2023 sourceExperimental — accuracy not yet measured
Section 5.4 (p. 10) investigates hallucinations in medical-domain translation, supporting the general point that hallucinations can occur in that setting. It does not examine chain-of-thought (CoT) or attribute hallucinations to misapplied medical terminology. It also reports that hallucination rates were not consistently higher for medical data than for the comparison data. (LLM, conf=0.82)
- Not matchedcites Ji et al., 2023
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.
- Also stated in: § A Analysis on the Addition of CoT · hedged
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“(ii) The demand for extensive domain knowledge (Schmidt and Rikers, 2007) and advanced reasoning skills (Liévin et al., 2022) makes eliciting medical expertise via simple prompting challenging (Kung et al., 2023;Singhal et al., 2023a).”
§ MEDAGENTS: Large Language Models as Collaborators for Zero-shot Medical Reasoning > 1 Introduction
What the sources say
- Not checked — no full text of the sourcecites Schmidt and Rikers, 2007 source
- Supportedcites Liévin et al., 2022 source
The Liévin et al. citation is attached to “advanced reasoning skills,” not to the later claim about simple prompting. The Introduction (p. 2) says realistic medical questions require advanced reasoning capabilities, and Appendix C (p. 23) describes USMLE questions as requiring strong problem-solving skills. (LLM, conf=0.93)
- Partly supportedcites Kung et al., 2023 sourceExperimental — accuracy not yet measured
The Introduction describes USMLE questions as testing extensive medical knowledge and clinical reasoning, and the Results report that ChatGPT gave some inaccurate answers to questions posed as natural user queries. That supports the idea that eliciting reliable answers is challenging. But the Discussion (“The rising accuracy of ChatGPT”; page unavailable) says ChatGPT approached the passing range without prompt tuning or training, so the paper does not support the broader suggestion that simple prompting prevents LLMs from handling medical tasks effectively. (LLM, conf=0.78)
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 cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Such failures in medical question-answering originate from a lack of domain knowledge (Harris, 2023;Kung et al., 2023;Tian et al., 2024) instead of reasoning rationale.”
§ MEDAGENTS: Large Language Models as Collaborators for Zero-shot Medical Reasoning > A Analysis on the Addition of CoT
What the sources say
- Not checked — no full text of the sourcecites Harris, 2023 source
- Partly supportedcites Kung et al., 2023 sourceExperimental — accuracy not yet measured
The Discussion (‘The rising accuracy of ChatGPT’; page unavailable) says inaccurate responses were driven primarily by missing information, supporting the domain-knowledge part of the claim. It does not establish that reasoning played no role: it also associates poorer accuracy with lower explanation concordance, and the Discussion (‘The potential for AI-assisted human learning in medical education’; page unavailable) says further failure-mode analysis is needed. (LLM, conf=0.88)
- Not matchedcites Tian et al., 2024
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
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 cited sources support only part of this claim — read each source’s verdict below.
In the paper
“For example, the results from Liévin et al. (2022) demonstrate that CoT improvements are significantly limited.”
§ MEDAGENTS: Large Language Models as Collaborators for Zero-shot Medical Reasoning > A Analysis on the Addition of CoT
What the sources say
- Partly supportedcites Liévin et al., 2022 sourceExperimental — accuracy not yet measured
The Discussion (p. 9) says single-sample CoT prompting was not competitive with direct prompting, supporting a limitation for that setting. But the Conclusion (p. 10) calls CoT a powerful tool that improved performance, and the Discussion (p. 10) reports gains from combining many CoTs. The unqualified claim that CoT improvements are significantly limited overstates the paper's finding. (LLM, conf=0.86)
boosted- 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
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
“We find that the optimal number of agents is relatively consistent across different datasets, pointing to its potential applicability to other datasets beyond those we test on.”
§ MEDAGENTS: Large Language Models as Collaborators for Zero-shot Medical Reasoning > 4.3 Number of agents
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.
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
“Notably, our proposed MEDAGENTS framework outperforms the zero-shot baseline methods by a large margin, indicating the effectiveness of our MEDAGENTS framework in real-world application scenarios.”
§ MEDAGENTS: Large Language Models as Collaborators for Zero-shot Medical Reasoning > 3.2 Main Results
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.
- Rests on an assumption: LLMs contain latent medical expertise that can be elicited through role-playing and multi-agent discussion.
- Rests on an assumption: The zero-shot setting is applicable in real-world scenarios.
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 Table 6 depict minor variance for different sizes of data with random removing, reinforcing the notion that large-scale experiment performance remains largely robust against the effect of domain changes.”
§ MEDAGENTS: Large Language Models as Collaborators for Zero-shot Medical Reasoning > 4.4 Domain Variation Study
plainn = 20- 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.
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 Table 4 demonstrate that the open-source methods fell short of the baseline in Table 2, which leads us to focus on the more effective methods.”
§ MEDAGENTS: Large Language Models as Collaborators for Zero-shot Medical Reasoning > 4.2 Comparison with Open-source Medical Models
boosted- 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
“Our key observation lies in that the performance improves significantly with the introduction of any number of expert agents compared to our baseline, thus verifying the consistent contribution of multiple expert agents.”
§ MEDAGENTS: Large Language Models as Collaborators for Zero-shot Medical Reasoning > 4.3 Number of agents
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
“Most notably, our approach achieves better performance under the zero-shot setting compared with the 5-shot baselines.”
§ MEDAGENTS: Large Language Models as Collaborators for Zero-shot Medical Reasoning > 1 Introduction
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.
- Rests on an assumption: The zero-shot setting is applicable in real-world scenarios.
- It may state two separate claims as one. 8
- Probably not a substantial claim. 6
- The quote only partly states this claim. 4
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 Table 3 show that all of these processes are non-trivial.”
§ MEDAGENTS: Large Language Models as Collaborators for Zero-shot Medical Reasoning > 4.1 Ablation Study
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
“Encouragingly, our proposed approach outperforms settings for both chain-of-thought (CoT) and self-consistency (SC) prompting methods.”
§ MEDAGENTS: Large Language Models as Collaborators for Zero-shot Medical Reasoning > 1 Introduction
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.
- Rests on an assumption: LLMs contain latent medical expertise that can be elicited through role-playing and multi-agent discussion.
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
“In our extensive examination of the MEDAGENTS framework, we discover the decent correctional capabilities of our framework.”
§ MEDAGENTS: Large Language Models as Collaborators for Zero-shot Medical Reasoning > 4.7 Correctional Capabilities and Interpretability
plain- 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 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
“*zero-shot setting Zero-shot Zero-shot CoT Zero-shot CoT + SC 61.3 54.3 44.3 MedAgents (Ours) 64.1 | 56.3 47.3 52.5 59.3 | 73.7 61.3 75.7 72.9 | 61.5 63.7 71.1 65.2 | | 76.2 61.9 75.1 77.7 | 63.6 53.2 68.8 69.8 | 74.0 66.0 76.0 79.0 | 75.4 62.1 82.3 82.1 | 75.0 67.8 65.3 58.3 75.7 70.9 78.5 72.1 |”
§ MEDAGENTS: Large Language Models as Collaborators for Zero-shot Medical Reasoning > Vote Result:
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.
The quote only partly states this claim.
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
“*zero-shot setting Zero-shot Zero-shot CoT Zero-shot CoT + SC 74.5 73.0 61.8 MedAgents (Ours) 83.7 | 69.0 69.0 70.1 74.8 | 76.2 71.0 75.3 76.8 | 78.5 82.1 80.0 83.5 | | 83.3 85.2 86.3 91.0 | 75.6 80.8 81.2 87.6 | 90.0 92.0 93.0 93.0 | 90.0 93.5 94.8 96.0 | 90.0 80.6 91.7 80.8 91.7 83.0 94.3 86.7 |”
§ MEDAGENTS: Large Language Models as Collaborators for Zero-shot Medical Reasoning > Vote Result:
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.
The quote only partly states this claim.
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
“MC framework GPT-4 (5 repetitions) 83.5 74.9”
§ MEDAGENTS: Large Language Models as Collaborators for Zero-shot Medical Reasoning > Additional information
plainn = 300- 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
“*zero-shot setting Zero-shot Zero-shot CoT Zero-shot CoT + SC 74.5 73.0 61.8 MedAgents (Ours) 83.7 | 69.0 69.0 70.1 74.8 | 76.2 71.0 75.3 76.8 | 78.5 82.1 80.0 83.5 | | 83.3 85.2 86.3 91.0 | 75.6 80.8 81.2 87.6 | 90.0 92.0 93.0 93.0 | 90.0 93.5 94.8 96.0 | 90.0 80.6 91.7 80.8 91.7 83.0 94.3 86.7 |”
§ MEDAGENTS: Large Language Models as Collaborators for Zero-shot Medical Reasoning > Vote Result:
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.
The quote only partly states this claim.
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
“*zero-shot setting Zero-shot Zero-shot CoT Zero-shot CoT + SC 61.3 54.3 44.3 MedAgents (Ours) 64.1 | 56.3 47.3 52.5 59.3 | 73.7 61.3 75.7 72.9 | 61.5 63.7 71.1 65.2 | | 76.2 61.9 75.1 77.7 | 63.6 53.2 68.8 69.8 | 74.0 66.0 76.0 79.0 | 75.4 62.1 82.3 82.1 | 75.0 67.8 65.3 58.3 75.7 70.9 78.5 72.1 |”
§ MEDAGENTS: Large Language Models as Collaborators for Zero-shot Medical Reasoning > Vote Result:
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.
The quote only partly states this 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
“This iterative refinement serves as a practical demonstration of our model's proficiency in rectifying errors, substantiating its interpretability and accuracy in complex medical reasoning tasks.”
§ MEDAGENTS: Large Language Models as Collaborators for Zero-shot Medical Reasoning > 4.7 Correctional Capabilities and Interpretability
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.
It may state two separate claims as one.
- Rests on an assumption: A unanimous report generated by iterative expert consultation best represents the collective expert knowledge and reasoning on the given medical problem.
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
“Different medical professionals offer depth and breadth in the analysis, transforming it into a comprehensive assessment.”
§ MEDAGENTS: Large Language Models as Collaborators for Zero-shot Medical Reasoning > Additional information
plaindirect- 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: fails — The paper may give no 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. 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
“Unlike these methods, our work emphasizes harnessing latent medical knowledge intrinsic to LLMs and improving reasoning in a training-free setting.”
§ MEDAGENTS: Large Language Models as Collaborators for Zero-shot Medical Reasoning > Example Interpretation
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.
- Rests on an assumption: LLMs contain latent medical expertise that can be elicited through role-playing and multi-agent discussion.
None of the sources cited for this claim could be matched to a paper in Crossref, so there is nothing to check it against.
In the paper
“In essence, the key to LLM-based multi-agent collaboration is the simulation of human activities such as role-playing (Wang et al., 2023d;Hong et al., 2023) and communication (Wu et al., 2023a;Qian et al., 2023;Li et al., 2023b,c).”
§ MEDAGENTS: Large Language Models as Collaborators for Zero-shot Medical Reasoning > 5.2 LLM-based Multi-agent Collaboration
What the sources say
- Not matchedcites Wang et al., 2023d
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites Hong et al., 2023
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites Wu et al., 2023a
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
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.
- Not matched
In the paper
“Although Med-PaLM (Singhal et al., 2023b) serves as a specialized medical LLM fine-tuned on the basis of PaLM 2, it is closed-sourced and not publicly accessible yet.”
§ MEDAGENTS: Large Language Models as Collaborators for Zero-shot Medical Reasoning > Vote Result:
What the sources say
- Not matchedcites Singhal et al., 2023b
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
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.
- Not matched
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“The process of selecting these domains is automated via prompting, and our manual inspection confirms the high relevance and quality of the selected domains.”
§ MEDAGENTS: Large Language Models as Collaborators for Zero-shot Medical Reasoning > 4.5 Agent Quantity Study
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
“Experimental results on nine datasets (MedQA, MedMCQA, PubMedQA, and six subtasks from MMLU) establish that our proposed MEDAGENTS framework excels at mining and harnessing the medical expertise within LLMs, as well as extending its reasoning abilities.”
§ MEDAGENTS: Large Language Models as Collaborators for Zero-shot Medical Reasoning > Abstract
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.
It may state two separate claims as one.
- Rests on an assumption: LLMs contain latent medical expertise that can be elicited through role-playing and multi-agent discussion.
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
“MedAgents leverages LLM-based agents in a role-playing setting that participate in a collaborative multi-round discussion, thereby enhancing LLM proficiency and reasoning capabilities.”
§ MEDAGENTS: Large Language Models as Collaborators for Zero-shot Medical Reasoning > Abstract
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.
- Rests on an assumption: LLMs contain latent medical expertise that can be elicited through role-playing and multi-agent discussion.
The paper relies on this without support.
LLMs contain latent medical expertise that can be elicited through role-playing and multi-agent discussion.
“Through such design, the expertise implicitly embedded within LLMs, or that the model has encountered during its training, which may not be readily accessible via traditional prompting, is effectively brought to the fore.”
The paper relies on this without support.
The zero-shot setting is applicable in real-world scenarios.
“Our work focuses on the zeroshot setting, which is applicable in realworld scenarios.”
The paper relies on this without support.
The multi-disciplinary consultation mechanism is common and effective in hospitals.
“Inspired by the multi-disciplinary consultation mechanism which is common and effective in hospitals, we are thus inspired to apply this mechanism to medical reasoning tasks through LLM-based multi-agent collaboration.”
The paper relies on this without support.
The authors are the first to propose a multi-agent framework within the medical domain.
“(i) To the best of our knowledge, we are the first to propose a multi-agent framework within the medical domain and explore how multi-agent communication within the medical setting can lead to a consensus decision, adding a novel dimension to the current literature on medical question answering.”
The paper relies on this without support.
A unanimous report generated by iterative expert consultation best represents the collective expert knowledge and reasoning on the given medical problem.
“Ultimately, after iterative refinement and consultation, a unanimous report is generated that best represents the collective expert knowledge and reasoning on the given medical problem.”
The paper relies on this without support.
The assembled LLM experts possess respective disciplinary backgrounds.
“The assembled LLM experts, possessing respective disciplinary backgrounds, engage in a discussion over the initial report, voicing agreements and disagreements.”
The paper relies on this without support.
An initial report synthesizing the analyses concisely represents the performed evaluations.
“An initial report synthesizing these analyses is then prepared to concisely represent the performed evaluations.”
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 cited source doesn’t support the claim as written.
In the citing paper
“MedAgents32 emphasizes QA improvement through agent collaboration but does not simulate patient interactions or decision-making processes.”
From AgentClinic: a multimodal benchmark for tool-using clinical AI agents (2026) · cites this paper as [32]
Checked by a reasoning model — may be wrong.
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
“Coding agents have also been applied to offensive security [11, 37, 60], theorem proving [44], and clinical tasks [38, 43, 52].”
From SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering (2024) · cites this paper as [43]
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
“Recent advancements in multi-agent systems have shown promising results in medical decision-making and diagnosis.”
From Enhancing diagnostic capability with multi-agents conversational large language models (2025) · cites this paper as [25]
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
“Current studies on multi-agent architectures have largely emphasized their enhanced text understanding capabilities, particularly in generative tasks62–65.”
From MoMA: a mixture-of-multimodal-agents architecture for enhancing clinical prediction modelling (2025) · cites this paper as [64]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“For example, MEDAGENTS is an innovative multi-disciplinary collaboration framework that enhances large language models’ performance in zero-shot medical question answering25.”
From Enhancing diagnostic capability with multi-agents conversational large language models (2025) · cites this paper as [25]
The check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“However, they have yet to demonstrate whether such architectures can effectively leverage their text understanding abilities to improve prediction performance in classification tasks involving multimodal EHR data, which is a critical need in clinical applications.”
From MoMA: a mixture-of-multimodal-agents architecture for enhancing clinical prediction modelling (2025) · cites this paper as [64]
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ MEDAGENTS: Large Language Models as Collaborators for Zero-shot Medical Reasoning > Vote Result:
“The final report suggested Tympanometry as the best medication for the patient's condition.”
§ MEDAGENTS: Large Language Models as Collaborators for Zero-shot Medical Reasoning > Question: Eustachian tube function is best assessed by: (A) Tympanometry (B) VEMP (C) Rhinomanometry (D) Politzer tests
“Question: Eustachian tube function is best assessed by: (A) Tympanometry (B) VEMP (C) Rhinomanometry (D) Politzer tests”
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
BioBERT: a pre-trained biomedical language representation model for biomedical text mining
Lee, J., Yoon, W., Kim, S., Kim, D., Kim, S., So, C. H., & Kang, J. (2020) · Bioinformatics · doi:10.1093/bioinformatics/btz682 · Cited by 7,218
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
Kung, T. H., Cheatham, M., Medenilla, A., Sillos, C., De Leon, L., Elepaño, C., Madriaga, M., Aggabao, R., Diaz-Candido, G., Maningo, J., et al. (2023) · PLOS Digital Health · doi:10.1371/journal.pdig.0000198 · Cited by 4,115
Large language models encode clinical knowledge
Singhal, K., Azizi, S., Tu, T., Mahdavi, S. S., Wei, J., Chung, H. W., Scales, N., Tanwani, A., Cole-Lewis, H., Pfohl, S., et al. (2023) · Nature · doi:10.1038/s41586-023-06291-2 · Cited by 4,083
Large language models in medicine
Thirunavukarasu, A. J., Ting, D. S. J., Elangovan, K., Gutierrez, L., Tan, T. F., & Ting, D. S. W. (2023) · Nature Medicine · doi:10.1038/s41591-023-02448-8 · Cited by 3,944
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
Generative Agents: Interactive Simulacra of Human Behavior
Park, J. S., O'Brien, J., Cai, C. J., Morris, M. R., Liang, P., & Bernstein, M. S. (2023) · Proceedings of the 36th Annual ACM Symposium on User Interface Software and Technology · doi:10.1145/3586183.3606763 · Cited by 2,187
Foundation models for generalist medical artificial intelligence
Moor, M., Banerjee, O., Abad, Z. S. H., Krumholz, H. M., Leskovec, J., Topol, E. J., & Rajpurkar, P. (2023) · Nature · doi:10.1038/s41586-023-05881-4 · Cited by 2,028
Large language models are zero-shot reasoners
Kojima, T., Gu, S. (., Reid, M., Matsuo, Y., & Iwasawa, Y. (2022) · Advances in Neural Information Processing Systems 35 · doi:10.52202/068431-1613 · Cited by 2,013
A survey on large language model based autonomous agents
Wang, L., Ma, C., Feng, X., Zhang, Z., Yang, H., Zhang, J., Chen, Z., Tang, J., Chen, X., Lin, Y., et al. (2024) · Frontiers of Computer Science · doi:10.1007/s11704-024-40231-1 · Cited by 1,941
Self-citationA Survey on Large Language Model based Autonomous Agents
Wang, L., Ma, C., Feng, X., Zhang, Z., Yang, H., Zhang, J., Chen, Z., Tang, J., Chen, X., Lin, Y., et al. (2023) · arXiv · doi:10.48550/arxiv.2308.11432 · Cited by 1,925
Self-citationToward expert-level medical question answering with large language models
Singhal, K., Tu, T., Gottweis, J., Sayres, R., Wulczyn, E., Amin, M., Hou, L., Clark, K., Pfohl, S. R., Cole-Lewis, H., et al. (2025) · Nature Medicine · doi:10.1038/s41591-024-03423-7 · Cited by 1,734
Sparks of Artificial General Intelligence: Early experiments with GPT-4
Bubeck, S., Chandrasekaran, V., Eldan, R., Gehrke, J., Horvitz, E., Kamar, E., Lee, P., Lee, Y. T., Li, Y., Lundberg, S., et al. (2023) · arXiv · doi:10.48550/arxiv.2303.12712 · Cited by 1,526
On Faithfulness and Factuality in Abstractive Summarization
Maynez, J., Narayan, S., Bohnet, B., & McDonald, R. (2020) · arXiv · doi:10.48550/arxiv.2005.00661 · Cited by 1,185
On Faithfulness and Factuality in Abstractive Summarization
Maynez, J., Narayan, S., Bohnet, B., & McDonald, R. (2020) · Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/2020.acl-main.173 · Cited by 1,185
Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debate
Liang, T., He, Z., Jiao, W., Wang, X., Wang, Y., Wang, R., Yang, Y., Shi, S., & Tu, Z. (2024) · Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2024.emnlp-main.992 · Cited by 1,062
LLaVA-Med: Training a Large Language-and-Vision Assistant for Biomedicine in One Day
Li, C., Wong, C., Zhang, S., Usuyama, N., Liu, H., Yang, J., Naumann, T., Poon, H., & Gao, J. (2023) · Advances in Neural Information Processing Systems 36 · doi:10.52202/075280-1240 · Cited by 986
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
Self-Consistency Improves Chain of Thought Reasoning in Language Models
Wang, X., Wei, J., Schuurmans, D., Le, Q., Chi, E., Narang, S., Chowdhery, A., & Zhou, D. (2022) · arXiv · doi:10.48550/arxiv.2203.11171 · Cited by 829
Jin, D., Pan, E., Oufattole, N., Weng, W.-H., Fang, H., & Szolovits, P. (2021) · Applied Sciences · doi:10.3390/app11146421 · Cited by 718
How expertise develops in medicine: knowledge encapsulation and illness script formation
Schmidt, H. G., & Rikers, R. M. J. P. (2007) · Medical Education · doi:10.1111/j.1365-2923.2007.02915.x · Cited by 619
ReAct: Synergizing Reasoning and Acting in Language Models
Yao, S., Zhao, J., Yu, D., Du, N., Shafran, I., Narasimhan, K., & Cao, Y. (2022) · arXiv · doi:10.48550/arxiv.2210.03629 · Cited by 608
LLaVA-Med: Training a Large Language-and-Vision Assistant for Biomedicine in One Day
Li, C., Wong, C., Zhang, S., Usuyama, N., Liu, H., Yang, J., Naumann, T., Poon, H., & Gao, J. (2023) · arXiv · doi:10.48550/arxiv.2306.00890 · Cited by 590
Graph of Thoughts: Solving Elaborate Problems with Large Language Models
Besta, M., Blach, N., Kubicek, A., Gerstenberger, R., Podstawski, M., Gianinazzi, L., Gajda, J., Lehmann, T., Niewiadomski, H., Nyczyk, P., et al. (2023) · arXiv · doi:10.48550/arxiv.2308.09687 · Cited by 556
Graph of Thoughts: Solving Elaborate Problems with Large Language Models
Besta, M., Blach, N., Kubicek, A., Gerstenberger, R., Podstawski, M., Gianinazzi, L., Gajda, J., Lehmann, T., Niewiadomski, H., Nyczyk, P., et al. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i16.29720 · Cited by 556
Chameleon: Plug-and-Play Compositional Reasoning with Large Language ModelsPreprint
Lu, P., Peng, B., Cheng, H., Galley, M., Chang, K.-W., Wu, Y., Zhu, S.-C., & Gao, J.-F. (2023) · Neural Information Processing Systems · doi:10.48550/arxiv.2304.09842 · Cited by 537
Capabilities of GPT-4 on Medical Challenge Problems
Nori, H., King, N., McKinney, S. M., Carignan, D., & Horvitz, E. (2023) · arXiv · doi:10.48550/arxiv.2303.13375 · Cited by 509
PubMedQA: A Dataset for Biomedical Research Question AnsweringPreprint
Jin, Q., Dhingra, B., Liu, Z., Cohen, W. W., & Lu, X. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1909.06146 · Cited by 487
Towards Generalist Biomedical AI
Tu, T., Azizi, S., Driess, D., Schaekermann, M., Amin, M., Chang, P.-C., Carroll, A., Lau, C., Tanno, R., Ktena, I., et al. (2023) · arXiv · doi:10.48550/arxiv.2307.14334 · Cited by 483
Towards Generalist Biomedical AI
Tu, T., Azizi, S., Driess, D., Schaekermann, M., Amin, M., Chang, P.-C., Carroll, A., Lau, C., Tanno, R., Ktena, I., et al. (2024) · NEJM AI · doi:10.1056/aioa2300138 · Cited by 483
Almanac — Retrieval-Augmented Language Models for Clinical Medicine
Zakka, C., Shad, R., Chaurasia, A., Dalal, A. R., Kim, J. L., Moor, M., Fong, R., Phillips, C., Alexander, K., Ashley, E., et al. (2024) · NEJM AI · doi:10.1056/aioa2300068 · Cited by 424
Evaluating large language models on medical evidence summarization
Tang, L., Sun, Z., Idnay, B., Nestor, J. G., Soroush, A., Elias, P. A., Xu, Z., Ding, Y., Durrett, G., Rousseau, J. F., et al. (2023) · npj Digital Medicine · doi:10.1038/s41746-023-00896-7 · Cited by 423
Opportunities and Challenges for ChatGPT and Large Language Models in Biomedicine and Health
Tian, S., Jin, Q., Yeganova, L., Lai, P.-T., Zhu, Q., Chen, X., Yang, Y., Chen, Q., Kim, W., Comeau, D. C., et al. (2023) · Briefings in Bioinformatics · doi:10.1093/bib/bbad493 · Cited by 359
Chameleon: Plug-and-Play Compositional Reasoning with Large Language Models
Lu, P., Peng, B., Cheng, H., Galley, M., Chang, K.-W., Wu, Y. N., Zhu, S. C., & Gao, J. (2023) · neural information processing systems · doi:10.52202/075280-1882 · Cited by 347
Can large language models reason about medical questions?
Liévin, V., Hother, C. E., Motzfeldt, A. G., & Winther, O. (2022) · arXiv · doi:10.48550/arxiv.2207.08143 · Cited by 343
Can large language models reason about medical questions?
Liévin, V., Hother, C. E., Motzfeldt, A. G., & Winther, O. (2024) · Patterns · doi:10.1016/j.patter.2024.100943 · Cited by 343
Towards Expert-Level Medical Question Answering with Large Language Models
Singhal, K., Tu, T., Gottweis, J., Sayres, R., Wulczyn, E., Hou, L., Clark, K., Pfohl, S., Cole-Lewis, H., Neal, D., et al. (2023) · arXiv · doi:10.48550/arxiv.2305.09617 · Cited by 338
CAMEL: Communicative Agents for "Mind" Exploration of Large Language Model Society
Li, G., Hammoud, H. A. A. K., Itani, H., Khizbullin, D., & Ghanem, B. (2023) · arXiv · doi:10.48550/arxiv.2303.17760 · Cited by 315
Camel: Communicative agents for" mind" exploration of large language model society
Li, G., Hammoud, H., Itani, H., Khizbullin, D., & Ghanem, B. (2023) · Advances in Neural Information Processing Systems 36 · doi:10.52202/075280-2264 · Cited by 314
Measuring Massive Multitask Language Understanding
Hendrycks, D., Burns, C., Basart, S., Zou, A., Mazeika, M., Song, D., & Steinhardt, J. (2020) · arXiv · doi:10.48550/arxiv.2009.03300 · Cited by 291
Self-citationRoleLLM: Benchmarking, Eliciting, and Enhancing Role-Playing Abilities of Large Language ModelsPreprint
Wang, Z. M., Peng, Z., Que, H., Liu, J., Zhou, W., Wu, Y., Guo, H., Gan, R., Ni, Z., Yang, J., et al. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2310.00746 · Cited by 291
Self-citationMetaGPT: Meta Programming for A Multi-Agent Collaborative Framework
Hong, S., Zhuge, M., Chen, J., Zheng, X., Cheng, Y., Zhang, C., Wang, J., Wang, Z., Yau, S. K. S., Lin, Z., et al. (2023) · arXiv · doi:10.48550/arxiv.2308.00352 · Cited by 214
BioMedLM: A 2.7B Parameter Language Model Trained On Biomedical TextPreprint
Bolton, E., Venigalla, A., Yasunaga, M., Hall, D., Xiong, B., Lee, T., Daneshjou, R., Frankle, J., Liang, P., Carbin, M., et al. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2403.18421 · Cited by 211
BioMedGPT: An Open Multimodal Large Language Model for BioMedicine
Luo, Y., Zhang, J., Fan, S., Yang, K., Hong, M., Wu, Y., Qiao, M., & Nie, Z. (2024) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2024.3505955 · Cited by 195
AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation
Wu, Q., Bansal, G., Zhang, J., Wu, Y., Li, B., Zhu, E., Jiang, L., Zhang, X., Zhang, S., Liu, J., et al. (2023) · arXiv · doi:10.48550/arxiv.2308.08155 · Cited by 193
Self-citationTheory of Mind for Multi-Agent Collaboration via Large Language Models
Li, H., Yu, C. H. A., Stepputtis, S., Campbell, J., Hughes, D., Lewis, C., & Sycara, K. P. (2023) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2023.emnlp-main.13 · Cited by 192
Jin, Q., Yang, Y., Chen, Q., & Lu, Z. (2024) · Bioinformatics · doi:10.1093/bioinformatics/btae075 · Cited by 176
Evaluation of the performance of GPT-3.5 and GPT-4 on the Polish Medical Final Examination
Rosoł, M., Gąsior, J. S., Łaba, J., Korzeniewski, K., & Młyńczak, M. (2023) · Scientific Reports · doi:10.1038/s41598-023-46995-z · Cited by 164
Analysis of large-language model versus human performance for genetics questions
Duong, D., & Solomon, B. D. (2024) · European Journal of Human Genetics · doi:10.1038/s41431-023-01396-8 · Cited by 159
MedMCQA : A Large-scale Multi-Subject Multi-Choice Dataset for Medical domain Question Answering
Pal, A., Umapathi, L. K., & Sankarasubbu, M. (2022) · arXiv · doi:10.48550/arxiv.2203.14371 · Cited by 157
OpenAgents: An Open Platform for Language Agents in the WildPreprint
Xie, T., Zhou, F., Cheng, Z., Shi, P. J., Weng, L., Liu, Y., Hua, T. J., Zhao, J., Liu, Q., Liu, C., et al. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2310.10634 · Cited by 153
Med-HALT: Medical Domain Hallucination Test for Large Language Models
Pal, A., Umapathi, L. K., & Sankarasubbu, M. (2023) · arXiv · doi:10.48550/arxiv.2307.15343 · Cited by 152
Med-HALT: Medical Domain Hallucination Test for Large Language Models
Pal, A., Umapathi, L. K., & Sankarasubbu, M. (2023) · Proceedings of the 27th Conference on Computational Natural Language Learning (CoNLL) · doi:10.18653/v1/2023.conll-1.21 · Cited by 152
Large Language Models Can Be Easily Distracted by Irrelevant Context
Shi, F., Chen, X., Misra, K., Scales, N., Dohan, D., Chi, E., Schärli, N., & Zhou, D. (2023) · arXiv · doi:10.48550/arxiv.2302.00093 · Cited by 134
DISC-MedLLM: Bridging General Large Language Models and Real-World Medical ConsultationPreprint
Bao, Z., Chen, W., Xiao, S., Ren, K., Wu, J., Zhong, C., Peng, J., Huang, X., & Wei, Z. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2308.14346 · Cited by 130
Knowledge-Augmented Reasoning Distillation for Small Language Models in Knowledge-Intensive TasksPreprint
Kang, M., Lee, S., Baek, J., et al. (2023) · Neural Information Processing Systems · doi:10.48550/arxiv.2305.18395 · Cited by 128
Hallucinations in Large Multilingual Translation Models
Guerreiro, N. M., Alves, D. M., Waldendorf, J., Haddow, B., Birch, A., Colombo, P., & Martins, A. F. T. (2023) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00615 · Cited by 124
Augmenting Black-box LLMs with Medical Textbooks for Biomedical Question Answering
Wang, Y., Ma, X., & Chen, W. (2024) · Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP) · doi:10.18653/v1/2024.findings-emnlp.95 · Cited by 93
Theory of Mind for Multi-Agent Collaboration via Large Language Models
Li, H., Chong, Y. Q., Stepputtis, S., Campbell, J., Hughes, D., Lewis, M., & Sycara, K. (2023) · arXiv · doi:10.48550/arxiv.2310.10701 · Cited by 82
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. (2023) · arXiv · doi:10.48550/arxiv.2307.07924 · Cited by 81
Wang, Z., Mao, S., Wu, W., Ge, T., Wei, F., & Ji, H. (2024) · Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers) · doi:10.18653/v1/2024.naacl-long.15 · Cited by 74
Large Language Models Answer Medical Questions Accurately, but Can’t Match Clinicians’ Knowledge
Harris, E. (2023) · JAMA · doi:10.1001/jama.2023.14311 · Cited by 70
PMC-VQA: Visual Instruction Tuning for Medical Visual Question Answering
Zhang, X., Wu, C., Zhao, Z., Lin, W., Zhang, Y., Wang, Y., & Xie, W. (2023) · arXiv · doi:10.48550/arxiv.2305.10415 · Cited by 66
Self-citationExamining Inter-Consistency of Large Language Models Collaboration: An In-depth Analysis via Debate
Xiong, K., Ding, X., Cao, Y., Liu, T., & Qin, B. (2023) · Findings of the Association for Computational Linguistics: EMNLP 2023 · doi:10.18653/v1/2023.findings-emnlp.508 · Cited by 66
Summarizing, Simplifying, and Synthesizing Medical Evidence using GPT-3 (with Varying Success)
Shaib, C., Li, M., Joseph, S. A., Marshall, I. J., Li, J. J., & Wallace, B. (2023) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2023.acl-short.119 · Cited by 58
PaperQA: Retrieval-Augmented Generative Agent for Scientific Research
Lála, J., O'Donoghue, O., Shtedritski, A., Cox, S., Rodriques, S. G., & White, A. D. (2023) · arXiv · doi:10.48550/arxiv.2312.07559 · Cited by 55
Knowledge-Augmented Reasoning Distillation for Small Language Models in Knowledge-Intensive Tasks
Kang, M., Lee, S., Baek, J., Kawaguchi, K., & Hwang, S. J. (2023) · neural information processing systems · doi:10.52202/075280-2109 · Cited by 54
PharmacyGPT: The AI PharmacistPreprint
Liu, Z., Wu, Z., Hu, M., Zhao, B., Zhao, L., Zhang, T., Dai, H., Chen, X., Shen, Y., Li, S., et al. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2307.10432 · Cited by 45
SciBERT: A Pretrained Language Model for Scientific Text
Beltagy, I., Lo, K., & Cohan, A. (2019) · arXiv · doi:10.48550/arxiv.1903.10676 · Cited by 41
Self-citationAgentVerse: Facilitating Multi-Agent Collaboration and Exploring Emergent Behaviors
Chen, W., Su, Y., Zuo, J., Yang, C., Yuan, C., Chan, C.-M., Yu, H., Lu, Y., Hung, Y.-H., Qian, C., et al. (2023) · arXiv · doi:10.48550/arxiv.2308.10848 · Cited by 39
WebArena: A Realistic Web Environment for Building Autonomous Agents
Zhou, S., Xu, F. F., Zhu, H., Zhou, X., Lo, R., Sridhar, A., Cheng, X., Ou, T., Bisk, Y., Fried, D., et al. (2023) · arXiv · doi:10.48550/arxiv.2307.13854 · Cited by 31
BioMedGPT: Open Multimodal Generative Pre-trained Transformer for BioMedicinePreprint
Luo, Y., Zhang, J., Fan, S., et al. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2308.09442 · Cited by 29
InvestLM: A Large Language Model for Investment using Financial Domain Instruction Tuning
Yang, Y., Tang, Y., & Tam, K. Y. (2023) · arXiv · doi:10.48550/arxiv.2309.13064 · Cited by 24
AlpaCare:Instruction-tuned Large Language Models for Medical ApplicationPreprint
Zhang, X., Tian, C., Yang, X., Chen, L.-C., Li, Z., & Petzold, L. R. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2310.14558 · Cited by 21
Self-citationTang, X., Tran, A., Tan, J. T. C., & Gerstein, M. (2023) · Clinical Natural Language Processing Workshop (ClinicalNLP) · doi:10.18653/v1/2023.clinicalnlp-1.58 · Cited by 14
Self-citationMedChatZH: a Better Medical Adviser Learns from Better InstructionsPreprint
Tan, Y., Li, M., Huang, Z., Yu, H., & Fan, G. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2309.01114 · Cited by 13
Summarizing, Simplifying, and Synthesizing Medical Evidence Using GPT-3 (with Varying Success)Preprint
Shaib, C., Li, M., Joseph, S. A., et al. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2305.06299 · Cited by 4
Aligning Factual Consistency for Clinical Studies Summarization through Reinforcement Learning
Tang, X., Cohan, A., & Gerstein, M. (2023) · Clinical Natural Language Processing Workshop (ClinicalNLP) · doi:10.18653/v1/2023.clinicalnlp-1.7 · Cited by 4
Self-citationTang, X., Tran, A., Tan, J. T. C., et al. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2305.05001 · Cited by 2
Self-citationOpenAI, Achiam, J., Adler, S., Agarwal, S., Ahmad, L., Akkaya, I., Aleman, F. L., Almeida, D., Altenschmidt, J., Altman, S., et al. (2023) · arXiv · doi:10.48550/arxiv.2303.08774
BLOOM: A 176B-Parameter Open-Access Multilingual Language Model
Workshop, B., :, Scao, T. L., Fan, A., Akiki, C., Pavlick, E., Ilić, S., Hesslow, D., Castagné, R., Luccioni, A. S., et al. (2022) · arXiv · doi:10.48550/arxiv.2211.05100
Self-citation
Research works printed but not linked: 3
- Ziwei Ji, Nayeon Lee, Rita Frieske, Tiezheng Yu, Dan Su, Yan Xu, Etsuko Ishii, Ye Jin Bang, Andrea Madotto, and Pascale Fung. 2023. Survey of hallucination in natural language generation. ACM Computing Surveys, 55(12):1-38.
- PMLR.
- PMLR.
Cited by
Sorted by citations
A Survey of Large Language Models
Zhao, W. X., Zhou, K., Li, J., Tang, T., Dong, Z., Hou, Y., Zhang, B., Min, Y., Zhang, J., Liu, P., et al. (2026) · Frontiers of Computer Science · doi:10.1007/s11704-026-60308-3 · Cited by 2,093
Self-citationA Comprehensive Overview of Large Language Models
Naveed, H., Khan, A. U., Qiu, S., Saqib, M., Anwar, S., Usman, M., Akhtar, N., Barnes, N., & Mian, A. (2025) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3744746 · Cited by 1,257
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
A survey on LLM-based multi-agent systems: Workflow, infrastructure, and challenges
Li, X., Wang, S., Zeng, S., Wu, Y., & Yang, Y. (2024) · Vicinagearth · doi:10.1007/s44336-024-00009-2 · Cited by 404
Metadata
- Authors
- Tang, X., Zou, A., Zhang, Z., Li, Z., Zhao, Y., Zhang, X., Cohan, A., & Gerstein, M.
- Affiliations
- Shanghai Jiao Tong University (CN)
- Venue
- Findings of the Association for Computational Linguistics ACL 2024
- Published
- 2024
- Pages
- 599-621
- OpenAlex
- W4402670791
- Cited by
- 376
- References
- 33
- Citations by year
- 2026: 156 · 2025: 157 · 2024: 62
- FWCI
- 79.79
- Type
- conference paper
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- Locations
- Findings of the Association for Computational Linguistics: ACL (Findings of ACL) (publishedVersion, cc-by)
- arXiv (Cornell University) (submittedVersion, cc-by)
- arXiv (Cornell University) (cc-by)
- Topics
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Multimodal Machine Learning Applications
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Keywords
- large language models, LLM agents, role-playing agents, training-free framework, medical question answering, expert consultation, decision-making
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 2026-10-07, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-07 (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
