Pith. sign in

REVIEW 34 cited by

Language agents achieve superhuman synthesis of scientific knowledge

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2409.13740 v2 pith:L7XGLXUW submitted 2024-09-10 cs.CL cs.AIcs.IRphysics.soc-ph

Language agents achieve superhuman synthesis of scientific knowledge

classification cs.CL cs.AIcs.IRphysics.soc-ph
keywords scientificlanguageliteraturepaperqa2tasksagentsmodelresearch
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

Language models are known to hallucinate incorrect information, and it is unclear if they are sufficiently accurate and reliable for use in scientific research. We developed a rigorous human-AI comparison methodology to evaluate language model agents on real-world literature search tasks covering information retrieval, summarization, and contradiction detection tasks. We show that PaperQA2, a frontier language model agent optimized for improved factuality, matches or exceeds subject matter expert performance on three realistic literature research tasks without any restrictions on humans (i.e., full access to internet, search tools, and time). PaperQA2 writes cited, Wikipedia-style summaries of scientific topics that are significantly more accurate than existing, human-written Wikipedia articles. We also introduce a hard benchmark for scientific literature research called LitQA2 that guided design of PaperQA2, leading to it exceeding human performance. Finally, we apply PaperQA2 to identify contradictions within the scientific literature, an important scientific task that is challenging for humans. PaperQA2 identifies 2.34 +/- 1.99 contradictions per paper in a random subset of biology papers, of which 70% are validated by human experts. These results demonstrate that language model agents are now capable of exceeding domain experts across meaningful tasks on scientific literature.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 34 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. AutoResearchBench: Benchmarking AI Agents on Complex Scientific Literature Discovery

    cs.AI 2026-04 accept novelty 8.0

    AutoResearchBench is a new benchmark showing top AI agents achieve under 10% success on complex scientific literature discovery tasks that demand deep comprehension and open-ended search.

  2. Probing the Critical Point (CritPt) of AI Reasoning: a Frontier Physics Research Benchmark

    cs.AI 2025-09 unverdicted novelty 8.0

    CritPt benchmark shows state-of-the-art LLMs reach only 5.7% average accuracy on full-scale unpublished physics research tasks, rising to about 10% with coding tools.

  3. Agentic generation of verifiable rules for deterministic, self-expanding reaction classification

    cs.AI 2026-07 unverdicted novelty 7.0

    Multi-agent LLMs generate and verify 14,073 deterministic reaction rules from 665,901 patents, enabling 97.7% classification of unseen reactions with finer resolution than fixed proprietary systems.

  4. NatureBench: Can Coding Agents Match the Published SOTA of Nature-Family Papers?

    cs.CL 2026-06 unverdicted novelty 7.0

    NatureBench evaluates ten frontier AI coding agents on 90 tasks from Nature papers under web-search-disabled conditions and finds the strongest agent surpasses published SOTA on only 17.8% of tasks, succeeding mainly ...

  5. Self-Driving Datasets: From 20 Million Papers to Nuanced Biomedical Knowledge at Scale

    cs.LG 2026-05 conditional novelty 7.0

    Starling, a multi-agent LLM system, extracts ~6.3 million nuanced structured records from PubMed across six tasks with reported error rates of 0.6-7.7%, lower than several curated databases.

  6. Self-Driving Datasets: From 20 Million Papers to Nuanced Biomedical Knowledge at Scale

    cs.LG 2026-05 conditional novelty 7.0

    Starling uses LLMs and agents to turn 22.5M PubMed papers into 6.3M nuanced structured records across six tasks with 0.6-7.7% frontier-model rejection rates, lower than error rates on existing curated databases.

  7. Kosmos: An AI Scientist for Autonomous Discovery

    cs.AI 2025-11 unverdicted novelty 7.0

    Kosmos is an AI scientist that maintains coherence over hundreds of agent steps via a shared world model, executes thousands of code lines and reads thousands of papers per run, and produces traceable reports with 79....

  8. OrchestrXR: A Multi-Agent System for Idea-to-Prototype XR Study Authoring

    cs.HC 2026-07 unverdicted novelty 6.0

    OrchestrXR uses multi-agent orchestration with structured schemas to generate Unity XR study prototypes from ideas, supported by a user study with 12 researchers indicating effective support and intent preservation.

  9. Multi-Turn Agentic Scientific Literature Search via Workflow Induction

    cs.CL 2026-07 unverdicted novelty 6.0

    PaperPilot induces executable DAG workflows for multi-turn literature search and trains via imitation plus preference optimization, raising Hit@5 from 58.0 to 77.0 over a baseline agent.

  10. One Reflection Is Not Enough: Self-Correcting Autonomous Research via Multi-Hypothesis Failure Attribution

    cs.AI 2026-06 unverdicted novelty 6.0

    SAGE with MHFA improves failure recovery in autonomous research agents, raising metrics-bearing outputs from 42% to 92% on a 12-topic benchmark versus single-reflection baselines.

  11. Thinking Like a Scientist? A Structural Study of LLM-Generated Research Methods

    cs.CL 2026-06 unverdicted novelty 6.0

    LLMs given only research questions from 1000 arXiv CS papers recommend a narrower set of methods than the original papers, with effective model-entity diversity dropping from 1232 to 59-96 and stronger agreement among...

  12. PaperFlow: Profiling, Recommending, and Adapting Across Daily Paper Streams

    cs.IR 2026-06 unverdicted novelty 6.0

    PaperFlow proposes a Profiling-Recommending-Adapting framework for longitudinal scientific paper recommendation and evaluates it on a new user-day benchmark with 24 simulated users, outperforming five baselines in ran...

  13. Self-Driving Datasets: From 20 Million Papers to Nuanced Biomedical Knowledge at Scale

    cs.LG 2026-05 unverdicted novelty 6.0

    An LLM entity-tagging pipeline plus multi-agent system extracts ~6.3M nuanced records from 22.5M PubMed papers across six tasks with lower measured error than existing curated databases.

  14. Effects of Generative AI Errors on User Reliance Across Task Difficulty

    cs.CY 2026-04 unverdicted novelty 6.0

    Higher generative AI error rates reduce user reliance, but task difficulty does not significantly moderate this effect.

  15. RExBench: Can coding agents autonomously implement AI research extensions?

    cs.CL 2025-06 unverdicted novelty 6.0

    RExBench is a new benchmark showing that LLM coding agents fail to autonomously implement most realistic research extensions to prior AI papers.

  16. Towards an AI co-scientist

    cs.AI 2025-02 unverdicted novelty 6.0

    A multi-agent AI system generates novel biomedical hypotheses that show promising experimental validation in drug repurposing for leukemia, new targets for liver fibrosis, and a bacterial gene transfer mechanism.

  17. Supervising the search process produces reliable and generalizable information-seeking agents

    cs.CL 2025-02 unverdicted novelty 6.0

    Process supervision via RAG-Gym produces more reliable and generalizable search agents, with gains driven by higher-quality queries on out-of-domain multi-hop tasks.

  18. Human--LLM Collaboration Is Transforming Complexity Metrics in Scientific Texts

    cs.CY 2026-06 unverdicted novelty 5.0

    Analysis of arXiv abstracts detects increased top-word turnover and flattening of LLM-style to complexity-metric relationships after 2022.

  19. Eliot: Interactively $\underline{E}$xploring Fast-Changing Scientific $\underline{Li}$terature Trends with $\underline{O}$nline Da$\underline{t}$a and Learning

    cs.IR 2026-05 unverdicted novelty 5.0

    Eliot is a query-time clustering and temporal visualization system for arXiv literature, evaluated via offline metrics on eight domains and a user survey showing 85% meaningful cluster labels.

  20. Vector RAG vs LLM-Compiled Wiki: A Preregistered Comparison on a Small Multi-Domain Research

    cs.CL 2026-05 conditional novelty 5.0

    A preregistered comparison on 24 papers found that an LLM-compiled wiki outperformed vector RAG on cross-document synthesis and citation accuracy but used more query tokens, with no system best across all metrics.

  21. Evidence-Grounded Frontier Mapping and Agentic Hypothesis Generation in Nanomedicine

    cs.AI 2026-05 unverdicted novelty 5.0

    pArticleMap combines article embeddings, graph-based frontier extraction, and agentic LLMs to map nanomedicine literature and generate hypotheses, achieving 10.8% gold recovery and 61% future-neighborhood rate in retr...

  22. CVEvolve: Autonomous Algorithm Discovery for Unstructured Scientific Data Processing

    cs.AI 2026-05 unverdicted novelty 5.0

    CVEvolve uses LLM agents with lineage-aware search to autonomously discover algorithms that outperform baselines on scientific image tasks including registration, peak detection, and segmentation.

  23. CVEvolve: Autonomous Algorithm Discovery for Unstructured Scientific Data Processing

    cs.AI 2026-05 unverdicted novelty 5.0

    CVEvolve is a zero-code LLM agent harness that autonomously discovers algorithms for scientific image tasks including registration, peak detection, and segmentation, reporting improvements over baselines via iterative...

  24. LLM-Oriented Information Retrieval: A Denoising-First Perspective

    cs.IR 2026-05 unverdicted novelty 5.0

    Denoising to maximize usable evidence density and verifiability is becoming the primary bottleneck in LLM-oriented information retrieval, conceptualized via a four-stage framework and addressed through a pipeline taxo...

  25. pAI/MSc: ML Theory Research with Humans on the Loop

    cs.AI 2026-04 unverdicted novelty 5.0

    pAI/MSc is a customizable multi-agent system that reduces human steering by orders of magnitude when turning a hypothesis into a literature-grounded, mathematically established, experimentally supported manuscript dra...

  26. EvoMaster: A Foundational Evolving Agent Framework for Agentic Science at Scale

    cs.AI 2026-04 unverdicted novelty 5.0

    EvoMaster is a self-evolving agent framework that achieves state-of-the-art results on scientific benchmarks by enabling iterative hypothesis refinement and knowledge accumulation across domains.

  27. Hephaestus: Toward a Cybersecurity AI Scientist

    cs.CR 2026-06 unverdicted novelty 4.0

    The paper proposes the Cybersecurity AI Scientist as a modular multi-agent architecture for automating cybersecurity research, distinguished by its focus on non-stationary threats and anchored in a four-zeros risk-tru...

  28. AutoResearch AI: Towards AI-Powered Research Automation for Scientific Discovery

    cs.AI 2026-05 unverdicted novelty 4.0

    A survey organizing AI-powered research automation into five workflow stages, defining AutoResearch and Vibe Research, and proposing five evaluation dimensions while noting domain-conditioned limits on autonomy.

  29. AI for Auto-Research: Roadmap & User Guide

    cs.AI 2026-05 unverdicted novelty 4.0

    The paper delivers a stage-by-stage roadmap for AI in research, showing reliable assistance in retrieval and tool tasks but fragility in novelty and judgment, advocating human-governed collaboration.

  30. LLM-Oriented Information Retrieval: A Denoising-First Perspective

    cs.IR 2026-05 unverdicted novelty 4.0

    Argues for a denoising-first paradigm in LLM-oriented information retrieval, framing challenges via a four-stage progression and providing a taxonomy of signal-to-noise optimization techniques across the pipeline.

  31. EvoMaster: A Foundational Evolving Agent Framework for Agentic Science at Scale

    cs.AI 2026-04 conditional novelty 4.0

    EvoMaster is an open-source agent framework whose iterative self-evolution loop achieves the highest average score on 9 of 10 scientific agent benchmarks compared to OpenHands, OpenClaw, and Codex.

  32. Multi-Dimensional Knowledge Profiling with Large-Scale Literature Database and Hierarchical Retrieval

    cs.CV 2026-01 unverdicted novelty 4.0

    Large-scale profiling of recent AI literature shows growth in safety, multimodal reasoning, and agent studies alongside stabilization in neural machine translation and graph methods.

  33. Evolving Roles of LLMs in Scientific Innovation: Assistant, Collaborator, Scientist, and Evaluator

    cs.DL 2025-07 unverdicted novelty 4.0

    The paper proposes a four-role framework for LLMs in scientific innovation and reviews methods, benchmarks, and limitations across Assistant, Collaborator, Scientist, and Evaluator roles.

  34. Position: Multimodal Large Language Models Can Significantly Advance Scientific Reasoning

    cs.CL 2025-02 unverdicted novelty 2.0

    Position paper claims multimodal LLMs can significantly advance scientific reasoning and proposes a four-stage roadmap plus challenges and suggestions.