Paper Citation Record · LEDGER

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents

As of 21 August 2026, this Paper Citation Record lists 49 of 49 outbound references and 2 inbound Pith citation observations for arXiv:2602.12984.

A citation records a reference. It does not transfer a finding from one paper to another.

pith.paper-citation-record.v1
2602.12984 v2

Coverage vector

measured 49 of 49 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-02T23:41:46.847782Z

measured 51 of 51 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-21T06:32:19.484+00:00

measured 2 of 2 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-06-27T09:34:09.347912Z

measured 0 of 1 external citation measurements

A source-named dated measurement, never combined with another source.

Source: pith, observed 2026-07-03T11:28:04.326064Z

Reference resolution

49 of 49 outbound references displayed

verified exact0
verified fuzzy0
unresolved48
parse uncertain0
malformed identifier1
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 74473242-8dee-46df-bf85-2a44e35342cf · outbound

This paper cites On Evaluation of Embodied Navigation Agents.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents On Evaluation of Embodied Navigation Agents

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:41.542179Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 950e1056-e214-47dd-8521-8f0845cb7eff · outbound

This paper cites System card: Claude opus 4 & claude sonnet 4.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents System card: Claude opus 4 & claude sonnet 4

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:41.626293Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation dc4b5753-9acd-4630-802b-d7c3fcf6e0a1 · outbound

This paper cites Claude sonnet 4.5 system card.https://www.anthropic.com/claude-sonne t-4-5-system-card, 2025.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents Claude sonnet 4.5 system card.https://www.anthropic.com/claude-sonne t-4-5-system-card, 2025

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:41.760033Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3e40eec4-fa03-4bb0-bf45-28bb5653ea85 · outbound

This paper cites SciMaster: Towards General-Purpose Scientific AI Agents, Part I. X-Master as Foundation: Can We Lead on Humanity's Last Exam?.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents SciMaster: Towards General-Purpose Scientific AI Agents, Part I. X-Master as Foundation: Can We Lead on Humanity's Last Exam?

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:41.833775Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4e29a8cc-7b09-435b-94b2-a796665c3b17 · outbound

This paper cites DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:41.937664Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3dc69148-fa01-465f-bef5-f6d5a7a2240b · outbound

This paper cites SuperGPQA: Scaling LLM Evaluation across 285 Graduate Disciplines.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents SuperGPQA: Scaling LLM Evaluation across 285 Graduate Disciplines

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:42.011986Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1ca25276-d045-401e-a83c-c106fa18b3ee · outbound

This paper cites RBench-V: A Primary Assessment for Visual Reasoning Models with Multi-modal Outputs.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents RBench-V: A Primary Assessment for Visual Reasoning Models with Multi-modal Outputs

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:42.127992Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b0663a81-39ac-4ee8-b7e8-d3f351d37055 · outbound

This paper cites R-Bench: Graduate-level Multi-disciplinary Benchmarks for LLM & MLLM Complex Reasoning Evaluation.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents R-Bench: Graduate-level Multi-disciplinary Benchmarks for LLM & MLLM Complex Reasoning Evaluation

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:42.234154Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4976358a-5bc9-445a-bfb7-55cb644d8af7 · outbound

This paper cites Discoveryworld: A virtual environ- ment for developing and evaluating automated scientific discovery agents.Advances in Neural Information Processing Systems, 37:10088–10116, 2024.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents Discoveryworld: A virtual environ- ment for developing and evaluating automated scientific discovery agents.Advances in Neural Information Processing Systems, 37:10088–10116, 2024

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:42.555326Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a84e9cb5-7ffb-4bf3-afd3-a6f9e5cf63c3 · outbound

This paper cites R-Bench: Graduate-level Multi-disciplinary Benchmarks for LLM & MLLM Complex Reasoning Evaluation.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents R-Bench: Graduate-level Multi-disciplinary Benchmarks for LLM & MLLM Complex Reasoning Evaluation

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:42.420473Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2188affb-e4be-4da2-934d-62c4c193b34b · outbound

This paper cites AgentBench: Evaluating LLMs as Agents.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents AgentBench: Evaluating LLMs as Agents

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:42.810611Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 03a0ca64-dbeb-45e7-b0fd-21cb65174b20 · outbound

This paper cites Search-o1: Agentic Search-Enhanced Large Reasoning Models.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents Search-o1: Agentic Search-Enhanced Large Reasoning Models

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:42.720642Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2adc9e63-4b74-4682-9bcc-435936b1d7d0 · outbound

This paper cites SciAgent: Tool-augmented Language Models for Scientific Reasoning.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents SciAgent: Tool-augmented Language Models for Scientific Reasoning

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:43.110460Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e2644418-bd9b-4aeb-916e-a46e51f3f435 · outbound

This paper cites Learn to explain: Multimodal reasoning via thought chains for sciencequestionanswering.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents Learn to explain: Multimodal reasoning via thought chains for sciencequestionanswering

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:42.972355Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9df05970-bd16-4978-bedd-dafad7ef3e9d · outbound

This paper cites Patil, Huanzhi Mao, Fanjia Yan, Charlie Cheng-Jie Ji, Vishnu Suresh, Ion Stoica, and Joseph E.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents Patil, Huanzhi Mao, Fanjia Yan, Charlie Cheng-Jie Ji, Vishnu Suresh, Ion Stoica, and Joseph E

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:43.489952Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4c320aa6-47e9-46bd-84c8-a3e4cfb63e72 · outbound

This paper cites GPT-4 Technical Report.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents GPT-4 Technical Report

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:43.244893Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9d12ad95-afbf-4ad3-8b9d-dc75de34e558 · outbound

This paper cites GPQA: A Graduate-Level Google-Proof Q&A Benchmark.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents GPQA: A Graduate-Level Google-Proof Q&A Benchmark

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:43.849352Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 13636209-cf9b-4ac8-aa01-4d39739d02c0 · outbound

This paper cites OpenAI GPT-5 System Card.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents OpenAI GPT-5 System Card

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:43.964010Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8b829228-cf1d-4d46-8ee0-fac425ef68e2 · outbound

This paper cites ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:43.668704Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 51aa645c-fae6-4fa5-a76c-97b7b2b589f8 · outbound

This paper cites A Survey of AI for Materials Science: Foundation Models, LLM Agents, Datasets, and Tools.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents A Survey of AI for Materials Science: Foundation Models, LLM Agents, Datasets, and Tools

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:44.325186Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 822f1f30-8980-4930-92de-533dc8e382ca · outbound

This paper cites SciPy 1.0--Fundamental Algorithms for Scientific Computing in Python.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents SciPy 1.0--Fundamental Algorithms for Scientific Computing in Python

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:44.466323Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 06f3a84d-5ff8-4037-8741-988b642024a5 · outbound

This paper cites Gemini: A Family of Highly Capable Multimodal Models.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents Gemini: A Family of Highly Capable Multimodal Models

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:44.143270Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5da8c59e-ac14-4cad-88ec-70bf662a7fb7 · outbound

This paper cites From AI for science to agentic science: A survey on autonomous scientific discovery.CoRR, abs/2508.14111, 2025.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents From AI for science to agentic science: A survey on autonomous scientific discovery.CoRR, abs/2508.14111, 2025

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:44.760711Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e2328cb1-f3cd-4bd8-88e7-2a7dd8f8ff5f · outbound

This paper cites AgentGym: Evaluating and training large language model-based agents across diverse environments.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents AgentGym: Evaluating and training large language model-based agents across diverse environments

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:44.879166Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 18793b37-a88c-4d30-aeee-1d82db5fb1e6 · outbound

This paper cites SciBench: Evaluating College-Level Scientific Problem-Solving Abilities of Large Language Models.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents SciBench: Evaluating College-Level Scientific Problem-Solving Abilities of Large Language Models

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:44.622336Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 043f1a4b-b753-4a0b-9e06-afe4d804ee16 · outbound

This paper cites BMMR: A Large-Scale Bilingual Multimodal Multi-Discipline Reasoning Dataset.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents BMMR: A Large-Scale Bilingual Multimodal Multi-Discipline Reasoning Dataset

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:45.067017Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation dd321c8a-80c4-48f9-8e85-7e8113c85983 · outbound

This paper cites On the Tool Manipulation Capability of Open-source Large Language Models.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents On the Tool Manipulation Capability of Open-source Large Language Models

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:45.208041Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 521c8bd2-31d2-4ddf-bc5f-0bfa840077e2 · outbound

This paper cites AgentGym-RL: Training LLM Agents for Long-Horizon Decision Making through Multi-Turn Reinforcement Learning.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents AgentGym-RL: Training LLM Agents for Long-Horizon Decision Making through Multi-Turn Reinforcement Learning

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:44.971735Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2ac982ff-d9a9-42a2-b187-1956894abcec · outbound

This paper cites Qwen3 Technical Report.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents Qwen3 Technical Report

Reference 29

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:45.360064Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f315465b-7099-4a1d-8722-cbb6ef739bb4 · outbound

This paper cites ReAct: Synergizing Reasoning and Acting in Language Models.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents ReAct: Synergizing Reasoning and Acting in Language Models

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:45.410108Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1ad184a2-420a-4a78-be22-8aa42cfb9da2 · outbound

This paper cites Wang, Peifeng Ruan, Donghan Yang, Tao Wang, Guanghua Xiao, Xin Liu, Carl Yang, Yang Xie, and Wenqi Shi.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents Wang, Peifeng Ruan, Donghan Yang, Tao Wang, Guanghua Xiao, Xin Liu, Carl Yang, Yang Xie, and Wenqi Shi

Reference 31

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:45.299746Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a366daec-3d43-43b2-a6f9-54bebbd1f34a · outbound

This paper cites debug-gym: A Text-Based Environment for Interactive Debugging.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents debug-gym: A Text-Based Environment for Interactive Debugging

Reference 32

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:45.552764Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3022e7e4-6289-4b75-b337-7e9140606382 · outbound

This paper cites MMMU-Pro: A More Robust Multi-discipline Multimodal Understanding Benchmark.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents MMMU-Pro: A More Robust Multi-discipline Multimodal Understanding Benchmark

Reference 33

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:45.646682Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7a2a993a-c2e0-4d3b-825b-f272da28d61e · outbound

This paper cites URLhttps://arxiv.org/abs/24 06.12045.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents URLhttps://arxiv.org/abs/24 06.12045

Reference 34

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:45.464222Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 20337007-c0ef-4811-a432-d37b4836ab93 · outbound

This paper cites Sciinstruct: a self-reflective instruction annotated dataset for training scientific language models.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents Sciinstruct: a self-reflective instruction annotated dataset for training scientific language models

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:45.806856Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a3a7ae48-876a-4096-9503-ade0391bc9f1 · outbound

This paper cites The Landscape of Agentic Reinforcement Learning for LLMs: A Survey.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents The Landscape of Agentic Reinforcement Learning for LLMs: A Survey

Reference 36

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:45.930272Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5312309a-044c-49d6-bfc6-5dc8604930c6 · outbound

This paper cites GLM-4.5: Agentic, Reasoning, and Coding (ARC) Foundation Models.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents GLM-4.5: Agentic, Reasoning, and Coding (ARC) Foundation Models

Reference 37

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:45.766345Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a4e71bd1-acc7-45b5-8874-68bee738df54 · outbound

This paper cites Scientists’ first exam: Probing cognitive abilities of mllm via perception, understanding, and reasoning.arXiv preprint arXiv:2506.10521, 2025.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents Scientists’ first exam: Probing cognitive abilities of mllm via perception, understanding, and reasoning.arXiv preprint arXiv:2506.10521, 2025

Reference 38

Resolution: malformed identifier; no resolver link, observed 2026-08-02T23:41:46.142352Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7e2daef1-1a3a-45d7-9cdb-109302624488 · outbound

This paper cites WebArena: A Realistic Web Environment for Building Autonomous Agents.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents WebArena: A Realistic Web Environment for Building Autonomous Agents

Reference 40

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:46.044618Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation df752ca1-612f-4c9f-8f17-37053c441d98 · outbound

This paper cites an unresolved cited work.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents Unresolved cited work

Reference 42

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:46.239829Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b1a4de21-0e95-4915-a80c-9eeca4213bd3 · outbound

This paper cites •Standard Return:{’result’: main_value, ’metadata’: {...}}(e.g., units, status flags, data sources).

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents •Standard Return:{’result’: main_value, ’metadata’: {...}}(e.g., units, status flags, data sources)

Reference 43

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:46.339736Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e5ff4846-6c47-45d5-a582-582625dc238e · outbound

This paper cites •Type Hints:All tools must provide complete Python type hints for parameters and return values.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents •Type Hints:All tools must provide complete Python type hints for parameters and return values

Reference 44

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:46.446763Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a0a8b7f6-be47-432e-b27b-768bd2e50fac · outbound

This paper cites •Query:Retrieve hierarchical facts/records from external resources or local indices and return normalized fields for downstream steps.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents •Query:Retrieve hierarchical facts/records from external resources or local indices and return normalized fields for downstream steps

Reference 45

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:46.552473Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 128b0d44-45b7-4905-b6de-2b5d57aed577 · outbound

This paper cites must include X.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents must include X

Reference 46

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:46.629695Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 62b6cd3d-2f28-46bd-8a08-6aed599e8cc9 · outbound

This paper cites an unresolved cited work.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents Unresolved cited work

Reference 47

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:46.712280Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3848077d-eac6-431d-85a0-b01af5be0ee1 · outbound

This paper cites an unresolved cited work.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents Unresolved cited work

Reference 48

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:46.758337Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 039c8a1a-3706-4739-84ad-c9d4779a0ae3 · outbound

This paper cites Tool-round limit exceeded; stopped automatically.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents Tool-round limit exceeded; stopped automatically

Reference 49

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:46.847782Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6ac35669-a6d2-4d1a-adbe-3287f9bab8c2 · outbound

This paper cites GPT-4 Technical Report.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents GPT-4 Technical Report

Reference 774

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:43.364678Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 23f30786-a418-42b2-961e-5a0b4ebefc36 · outbound

This paper cites an unresolved cited work.

SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents Unresolved cited work

Reference 2025

Resolution: unresolved; no resolver link, observed 2026-08-02T23:41:41.688635Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation 34e2cafd-a6ce-4171-a250-d665cac90c96 · inbound

AI scientists produce results without reasoning scientifically cites this paper.

AI scientists produce results without reasoning scientifically SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents

Reference 5

Resolution: verified exact; arxiv_id, observed 2026-06-02T02:03:38.947361Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 4a5d332b-b3ee-410f-93d2-9b9884877e8d · inbound

Benchmarking AI Agents for Addressing Scientific Challenges Across Scales cites this paper.

Benchmarking AI Agents for Addressing Scientific Challenges Across Scales SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents

Reference 12

Resolution: verified exact; local_arxiv, observed 2026-07-03T11:28:04.327347Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.