Pith.

Artificial Intelligence · arXiv:2609.17523 v1 · submitted 15 Sep 2026 · 29 references

ScienceBuddy: Recursive-in-Recursive Self-Improvement for Interactive Scientific Agents

Shuhan Xue, Jianyuan Zhong, Ziyuan Nan, Wenbin Li, Zhaochen Yu, Jinchao Ding, Qiang Gao, Pengyu Zhan

show 5 more authors

Yuntong Zhang, Tian Cheng, Zhenfei Yin, Yingcheng Wu, Ling Yang

The paper says

“We introduce and release ScienceBuddy, an interactive scientific research workspace that brings continually improving scientific agents into researchers' everyday workflows.”

more of the abstract

ScienceBuddy supports researchers in carrying out scientific tasks while transforming their requests, feedback, and execution evidence into tasks and evaluation rubrics for continual learning. At its core is recursive-in-recursive self-improvement, a paradigm that couples harness evolution with model reinforcement learning: the inner recursion improves the harness with the model fixed, while the outer recursion trains the model under the improved harness. Harness evolution shapes training experience, and model learning creates new opportunities for harness adaptation. We present case studies of researcher interaction, harness refinement, and model learning, with the benchmark cases spanning four scientific task families. By releasing ScienceBuddy as a research product, we make this paradigm available to the scientific community and take a step toward discovery intelligence: scientific AI that advances through sustained collaboration with researchers and evolves alongside the research it supports. Website: http://science-buddy.io

from the abstract

Pith says

Machine review · deepseek-v4-flash · 17 Sep 2026 · what was read was not recorded · not checked by a person

It can be wrong. Challenge this review or reply to it.

Holds up, with conditions. Interesting architecture and solid engineering, but the central claim about nested recursion is undersupported by missing ablation controls. This line is wrong

3 major objections · 5 minor

How often objections from reviews like this one are right: not yet measured

Are you an author? What you can do.

Not checked: figures, code, what the 29 references say.

The paper says, at §4.2, Fig. 8

The review did not quote this passage.

Pith says: Objection 1 of 3 · major

The three-cycle experiment runs harness evolution and model RL together but never compares against a baseline that applies only harness adaptation across multiple cycles or only model RL across multiple cycles. The standalone experiments in §4.3 and §4.4 are conducted on different task sets and evaluation protocols (first-response accuracy vs. pass@4) and are not repeated over multiple cycles, so they cannot serve as multi-cycle ablations.

more of objection 1

The central novelty of the paper is the recursive-in-recursive coupling; without a controlled ablation, the evidence that the coupling sustains improvement beyond either component alone is inconclusive. The authors should either provide multi-cycle baselines for each component separately, or substantially weaken the claim to describe the full system as one plausible instantiation rather than a demonstrated necessity.

The paper says, at S1.2 (simulated user feedback)

The review did not quote this passage.

Pith says: Objection 2 of 3 · major

The harness evolution in both the coupled experiment (§4.2) and the harness-only experiment (§4.3) relies on simulated user feedback generated by a bounded user simulator (Qwen3.8-27B helper with a fixed reply set). The paper acknowledges this limitation but does not discuss how real researcher feedback might differ in distribution, noise, or informativeness, nor whether the bounded simulator biases the harness toward a narrow set of procedural patterns.

more of objection 2

A sensitivity analysis or an informal comparison with real interaction trajectories (e.g., from §4.1) would help assess how representative the simulated feedback is.

The paper says, at §5 (Related Work, p. 17)

The review did not quote this passage.

Pith says: Objection 3 of 3 · major

The paper states that 'the reflector remains fixed, so improved task performance does not imply that the improvement mechanism itself has become stronger.' This is an honest limitation, but it undercuts the term 'self-improvement.' The nested recursion is the paper's headline contribution; the reader should be told more explicitly what kind of self-improvement is and is not being demonstrated.

more of objection 3

Future work or discussion should address whether the fixed reflector is a temporary engineering choice or a principled limitation of the paradigm.

5 minor objections

The paper says, at Abstract

Pith says: minor objection 1 of 5

The abstract uses 'continually improving scientific agents' but does not clarify that the improvement is in task performance, not in the improvement mechanism itself. Rephrasing to 'continually improving scientific task performance' would prevent over-interpretation.

The paper says, at §2.2, Eq. (2)

Pith says: minor objection 2 of 5

ConstructRubric is called but not fully defined in the main text; the function relies on the full collaboration trajectory, reconstructed instruction, and assets. A brief note on what happens when the rubric derivation fails or produces conflicting criteria would improve clarity.

show 3 more

The paper says, at §4.2, Fig. 8 caption

Pith says: minor objection 3 of 5

The caption refers to 'three RinR cycles' but the figure legend uses stages 1, 2, 3. Align terminology (e.g., cycle 1, 2, 3) for consistency.

The paper says, at §S1.2, Table S2

Pith says: minor objection 4 of 5

The information-boundary table is helpful. The row for 'Scientific verifier' says it 'Requires output' – this should read 'Required output' for parallel phrasing with other rows.

The paper says, at §S1.3, Eq. (S1)–(S3)

Pith says: minor objection 5 of 5

The group-relative advantage definition (Eq. S1) uses std denominator with a small 𝛿. It would be helpful to state the value of 𝛿 used in the experiments.

What would settle it

Run a matched ablation: starting from the same initial (θ₀, H₀), execute three cycles of harness-only adaptation (inner recursion, 10 steps per cycle) without any model RL, and separately three cycles of model-only RL (20 updates per cycle) with a fixed harness. Evaluate both baselines on the same held-out test set used in §4.2. Compare their final accuracy to the full system’s 73.3%. If either baseline matches or exceeds that figure, the recursive-in-recursive coupling is not the cause of the improvement.

from the skeptic's reading

How the review went

4 readings by deepseek-v4-flash · 17 Sep 2026

  1. Reader holds up, with conditions confidence moderate
  2. Skeptic partly agrees 33 s · 22,020 tokens read, 3,063 written
  3. Editor wrote the verdict above 21 s · 23,431 tokens read, 611 written
  4. Referee major revision: 3 major objections, 5 minor 20 s · 22,398 tokens read, 1,871 written

Not run on this paper: the simulated authors' defence.

Pith did not check this review's objections against the paper.

The review in full

The review in brief, machine-written

This paper introduces a novel recursive-in-recursive framework for scientific agent self-improvement, coupling harness evolution with model RL. The system is well-described, the code is released, and the case studies show consistent improvement on held-out tasks (42.2% to 73.3% over three cycles). The standalone harness and model experiments each demonstrate that both components can improve individually.

The soft spot is load-bearing: the central novelty is the nested coupling, but the paper never compares the full system to repeated application of either component alone. The three-cycle experiment couples both, while the harness-only and model-only experiments use different tasks and metrics (first-response accuracy vs. pass@4) and are not repeated over multiple cycles. We cannot tell whether the coupling is necessary or whether simply running harness evolution or model RL alone over multiple cycles would match or exceed those gains. This weakness is explicit in the stress test and, on reading the paper, I agree it is not addressed.

The paper is transparent about other limitations (simulated user feedback, small 4B model, few cycles, fixed reflector), and those are honestly stated. But the key empirical claim that the nested recursion sustains improvement lacks the direct ablation control it needs. The idea is interesting and the infrastructure is real, but the evidence for the coupling is circumstantial.

For a researcher interested in agent architectures or continual learning, this paper provides a clear description and a practical system. But they should read it with a skeptical eye on the central claim. I would recommend conditional acceptance with a request for an ablation study that compares the full recursive system to at least one of the component-only baselines over matched cycles.

The strongest objection

The central novelty of ScienceBuddy is the recursive-in-recursive coupling of harness evolution and model RL. The headline three-cycle experiment (§4.2) runs both components together: each cycle does 10 harness-evolution steps followed by 20 RL updates. But the paper never compares this full system to a baseline that runs only harness adaptation across multiple cycles (without model updates) or only model RL across multiple cycles (with a fixed harness). The standalone harness-only (§4.3) and model-only (§4.4) experiments each show improvement, but they are conducted on different task sets and evaluation protocols (first-response accuracy vs. pass@4) and are not repeated over multiple cycles. Thus we cannot determine whether the coupling is necessary or whether a simpler repeated application of either component alone would yield comparable or better gains. If, for example, three cycles of harness-only adaptation achieve the same 73.3% test accuracy, the recursive-in-recursive claim would lose force. The paper acknowledges that the reflector remains fixed (§5, p.17: “improved task performance does not imply that the improvement mechanism itself has become stronger”), which limits the strength of self-improvement, but does not address this more fundamental ablation gap. This is load-bearing because the entire paradigm is defined by the nested recursion; without an ablation control, the evidence that the coupling sustains improvement across cycles is inconclusive.

The reader’s weakest assumption was the use of simulated user feedback, which is a separate limitation about external validity. My concern is about internal validity: the experiment does not isolate the effect of the nested recursion. Both are legitimate concerns, but mine targets the core claim that the coupling sustains improvement. Since the reader’s verdict was already CONDITIONAL and flagged the need for more realistic validation, adding this methodological point does not change the verdict—it strengthens the conditions under which the claim would be accepted. I recommend UNCHANGED because the concern does not fundamentally contradict the reader’s assessment but rather adds specificity about what additional evidence is needed.

The referee's report

If the central claim holds, the recursive-in-recursive paradigm could provide a principled mechanism for continual improvement of scientific AI agents through routine researcher collaboration. The coupling of procedural editing with model-level RL is a natural direction, and the release of a functional workspace with 224 tools across 22 modules could enable community experiments. The paper also demonstrates measurable gains on established benchmarks (LAB-Bench, Biomni-Eval1) with a small 4B model, which is encouraging for cost-sensitive deployment. However, the significance is currently limited by the reliance on simulated user feedback and the absence of an ablation that isolates the contribution of the recursive coupling itself.

Pith's reading of the paper (a machine's summary, not the authors')

ScienceBuddy is an interactive scientific workspace where a language-model agent improves by alternating two nested loops. In the inner loop, the agent's instructions and skills (the harness) are revised while the model remains frozen. In the outer loop, the model is updated via reinforcement learning under the improved harness, using task-specific rubric scores as rewards. Across three cycles of simulated researcher interaction, the agent's single-attempt test accuracy on four scientific task families rose from 42.2% to 73.3%, with only 2.2% regression. The paper argues this coupling of procedural and policy learning can sustain improvement as the system continues to interact with researchers.

What carries the argument. Recursive-in-recursive self-improvement: an inner recursion that keeps the task model fixed while a separate auxiliary model diagnoses execution failures and proposes bounded edits to the harness (instruction entries, reusable skills, and context settings), accepting candidates only when they improve a paired development evaluation; and an outer recursion that augments the task environment, collects fresh on-policy rollouts under the selected harness, and trains the model via group-relative policy optimization (GRPO) with rubric-based trajectory rewards.

Core claim. The central claim is that a scientific AI agent can continuously improve both its working procedures and its underlying capabilities by coordinating harness evolution and model reinforcement learning. The harness is refined by a fixed auxiliary model that diagnoses failures and proposes bounded edits to instructions and skills; the updated harness then guides fresh rollouts whose rubric rewards drive GRPO-based model training. The case studies show that harness adaptation alone raises first-response accuracy by 20 percentage points at fixed model weights, and model learning alone increases problem coverage by 19.5 percentage points under a fixed harness. The coupled three-cycle experiment demonstrates overall accuracy gains of 31.1 percentage points, suggesting that the two improvement mechanisms complement each other and do not plateau after a single round.

The premise it rests on. The harness-evolution case study relies on simulated user feedback generated by a bounded user simulator, not on real researcher interactions; if simulated feedback differs substantially from real behavior, the harness improvements may not transfer to actual scientific collaboration.

If the paper is right (Pith's inferences, not the authors')

Scientific AI agents can improve without full retraining by adapting how they present instructions and use tools, enabling cost-effective capability growth.

The methodology provides a template for building interactive agents in other domains that learn from user feedback, provided rubrics can be derived from collaboration.

The three-cycle experiment shows sustained improvement without saturation, suggesting potential for indefinite self-improvement as researcher interactions accumulate.

Problem-coverage gains (pass@4 from 48.3% to 67.8%) indicate that model learning can broaden the set of solvable tasks, not just sharpen accuracy on known ones.

The coupling of harness and model updates means that improvements in one component create new opportunities for the other, producing a compounding effect.

If the method transfers to real researcher interactions (not just simulators), it could turn routine scientific chat into a continuous training signal, reducing the need for manual data curation.

The recursive-in-recursive scheme effectively uses the agent's history as a self-generated curriculum for both procedures and policy, implying a general architecture for open-ended agent learning.

The fixed auxiliary model that proposes harness edits is a bottleneck; replacing it with a learned editor could make the inner loop itself improvable, though the paper explicitly avoids this to isolate the coupling.

Because rubrics are derived from collaboration but fixed during RL, the quality of rubric construction directly bounds the reward signal; poorly designed criteria could misguide training or miss meaningful improvements.

Check this review

Pith stored no record of this review's passes and took no fingerprint of it. Reviews made by code from 13 Nov 2026 on carry both.

How often reviews like this one are right

Reviews like this one: written by deepseek-v4-flash, where what was read was not recorded.

How often objections from reviews like this one are right: not yet measured

Every value above, in the review's public record (JSON)

Reviews by people

This review was written by a machine. No person has reviewed this paper on Pith yet. A scientist listed on Pith can review it for a fee, and their review is published here whatever it concludes.

Request a human review

How Pith checks papers · The full audit of this paper

The paper

Abstract

We introduce and release ScienceBuddy, an interactive scientific research workspace that brings continually improving scientific agents into researchers' everyday workflows. ScienceBuddy supports researchers in carrying out scientific tasks while transforming their requests, feedback, and execution evidence into tasks and evaluation rubrics for continual learning. At its core is recursive-in-recursive self-improvement, a paradigm that couples harness evolution with model reinforcement learning: the inner recursion improves the harness with the model fixed, while the outer recursion trains the model under the improved harness. Harness evolution shapes training experience, and model learning creates new opportunities for harness adaptation. We present case studies of researcher interaction, harness refinement, and model learning, with the benchmark cases spanning four scientific task families. By releasing ScienceBuddy as a research product, we make this paradigm available to the scientific community and take a step toward discovery intelligence: scientific AI that advances through sustained collaboration with researchers and evolves alongside the research it supports. Website: http://science-buddy.io

The paper's own sentences, arranged, kept beside the review and separate from it. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. No sentence in this record was written by a model.

abstract compiled · 2026-09-23

The authors' main claim, as the classifier picked it out 0.93

“By releasing ScienceBuddy as a research product, we make this paradigm available to the scientific community and take a step toward discovery intelligence: scientific AI that advances through sustained collaboration with researchers and evolves alongside the research it supports.”

How the main claim is stated: asserted 0.80

What the abstract itself states, as probabilities: claims first 0.07 · states limitation 0.07 · human participants 0.75 · controlled comparison 0.04 · randomized assignment 0.01 · states quantitative result 0.08 · compares with prior results 0.06 · mentions code or data availability 0.87

The abstract, sentence by sentence

This table labels the abstract only (arXiv listing abstract v1). The number is the probability the classifier gave the label.

SentenceRoleThe authors' own workStates the main conclusionStrength
We introduce and release ScienceBuddy, an interactive scientific research workspace that brings continually improving scientific agents into researchers' everyday workflows. contribution 1.00 0.91 0.16 asserted 1.00
ScienceBuddy supports researchers in carrying out scientific tasks while transforming their requests, feedback, and execution evidence into tasks and evaluation rubrics for continual learning. contribution 0.62 0.68 0.08 asserted 1.00
At its core is recursive-in-recursive self-improvement, a paradigm that couples harness evolution with model reinforcement learning: the inner recursion improves the harness with the model fixed, while the outer recursion trains the model under the improved harness. method 0.91 0.60 0.18 asserted 1.00
Harness evolution shapes training experience, and model learning creates new opportunities for harness adaptation. method 0.51 0.31 0.12 asserted 1.00
We present case studies of researcher interaction, harness refinement, and model learning, with the benchmark cases spanning four scientific task families. contribution 0.76 0.93 0.08 asserted 0.99
By releasing ScienceBuddy as a research product, we make this paradigm available to the scientific community and take a step toward discovery intelligence: scientific AI that advances through sustained collaboration with researchers and evolves alongside the research it supports. contribution 0.51 0.84 0.53 asserted 0.80
Website: http://science-buddy.io contribution 0.66 0.26 0.02 unclear 0.96

What kind of paper the classifier reports this is: method or tool 1.00 · evidence from observation or experiment 0.85

What was examined, and what was not

Sources read

  • arXiv listing abstract v1

Not examined

  • full text
  • figures and tables
  • references and citing sentences
  • relations to other papers

Levels run

  • P abstract

Not run

  • S sections
  • T body sentences
  • C citations
  • R cross-paper relations

How the labels were made

Compiled by classifier-1.13.0 via TypeSafe API, question bank abstract-p1-20260923, 2026-09-23: 1 request, 5,561 input tokens.

Every quoted sentence on this record is the authors'. Every label is from Pith's closed vocabulary, with the probability shown beside it. This record as JSON

References (29)
  1. [1] Harbor: Task Structure. https://www.harborframework.com/docs/tasks. Accessed September 9, 2026 on Pith
  2. [2] Lakshya A Agrawal, Shangyin Tan, Dilara Soylu, Noah Ziems, Rishi Khare, Krista Opsahl-Ong, Arnav Singhvi, Herumb Shandilya, Michael J Ryan, Meng Jiang, Christopher Potts, Koushik Sen, Alexandros G. Dimakis, Ion Stoica, Dan Klein, Matei Zaharia, and Omar Khattab. GEPA: Reflecti... on PithPith's reviewarXivdoi:10.48550/arxiv.2507.19457
  3. [3] Shraddha Barke, Arnav Goyal, Alind Khare, Avaljot Singh, Suman Nath, and Chetan Bansal. AgentRx: Diagnosing AI Agent Failures from Execution Trajectories.arXiv preprint arXiv:2602.02475, 2026. URLhttps://arxiv.org/abs/2602.02475 on PithPith's reviewarXiv
  4. [4] Liang Ding. AdaRubric: Task-Adaptive Rubrics for Reliable LLM Agent Evaluation and Reward Learning.arXiv preprint arXiv:2603.21362, 2026. URLhttps://arxiv.org/abs/2603.21362 on PithPith's reviewarXiv
  5. [5] Tianyu Fan and Chao Huang. HELIX: Model-Harness Co-evolution for Recursive Self-Improvement. arXiv preprint arXiv:2608.13951, 2026. URLhttps://arxiv.org/abs/2608.13951 on PithPith's reviewarXiv
  6. [6] Zhiyuan Fan, Tinghao Yu, Yuanjun Cai, Jiang Zhou, Jiangtao Guan, Jincheng Liu, Yun Yang, Dingxin Hu, Zhuo Han, Xing Wu, Feng Zhang, and Lilin Wang. Environment Evolution for Terminal Agents. arXiv preprint arXiv:2609.04128, 2026. URLhttps://arxiv.org/abs/2609.04128 on PithPith's reviewarXiv
  7. [7] Anisha Gunjal, Anthony Wang, Elaine Lau, Vaskar Nath, Yunzhong He, Bing Liu, and Sean Hendryx. Rubrics as Rewards: Reinforcement Learning Beyond Verifiable Domains.arXiv preprint arXiv:2507.17746, 2025. URLhttps://arxiv.org/abs/2507.17746 on PithPith's reviewarXiv
  8. [8] Prannay Hebbar, Yogendra Manawat, Samuel Verboomen, Alesia Ivanova, Selvam Palanimalai, Kunal Bhatia, and Vignesh Baskaran. SIA: Self Improving AI with Harness & Weight Updates.arXiv preprint arXiv:2605.27276, 2026. URLhttps://arxiv.org/abs/2605.27276 on PithPith's reviewarXiv
  9. [9] Kexin Huang, Serena Zhang, Hanchen Wang, Yuanhao Qu, Yingzhou Lu, Yusuf Roohani, Ryan Li, Lin Qiu, Junze Zhang, Yin Di, et al. Biomni: A General-Purpose Biomedical AI Agent.bioRxiv, 2025. doi: 18 ScienceBuddy: Recursive-in-Recursive Self-Improvement 10.1101/2025.05.30.656746. ... on Pithdoi:10.1101/2025.05.30.656746
  10. [10] Jon M. Laurent, Joseph D. Janizek, Michael Ruzo, Michaela M. Hinks, Michael J. Hammerling, Siddharth Narayanan, Manvitha Ponnapati, Andrew D. White, and Samuel G. Rodriques. LAB-Bench: Measuring Capabilities of Language Models for Biology Research.arXiv preprint arXiv:2407.10362, on PithPith's reviewarXiv
  11. [11] Yoonho Lee, Roshen Nair, Qizheng Zhang, Kangwook Lee, Omar Khattab, and Chelsea Finn. Meta- Harness: End-to-End Optimization of Model Harnesses.arXiv preprint arXiv:2603.28052, 2026. doi: 10.48550/arXiv.2603.28052. URLhttps://arxiv.org/abs/2603.28052 on PithPith's reviewarXivdoi:10.48550/arxiv.2603.28052
  12. [12] Haoyue Liu, Zhichao Wang, Yongxin Guo, Haoran Shou, and Xiaoying Tang. Adaptive Prompt Structure Factorization: A Framework for Self-Discovering and Optimizing Compositional Prompt Programs.arXiv preprint arXiv:2604.06699, 2026. URLhttps://arxiv.org/abs/2604.06699 on PithPith's reviewarXiv
  13. [13] Yuchen Ma, Yue Huang, Han Bao, Haomin Zhuang, Swadheen Shukla, Michel Galley, Xiangliang Zhang, and Stefan Feuerriegel. SkillGen: Verified Inference-Time Agent Skill Synthesis.arXiv preprint arXiv:2605.10999, 2026. URLhttps://arxiv.org/abs/2605.10999 on PithPith's reviewarXiv
  14. [14] Zhihong Shao, Peiyi Wang, Qihao Zhu, Runxin Xu, Junxiao Song, Xiao Bi, Haowei Zhang, Mingchuan Zhang, Y. K. Li, Y. Wu, and Daya Guo. DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models.arXiv preprint arXiv:2402.03300, 2024. URLhttps://arxiv.org/ ... on PithPith's reviewarXiv
  15. [16] Noah Shinn, Federico Cassano, Edward Berman, Ashwin Gopinath, Karthik Narasimhan, and Shunyu Yao. Reflexion: Language Agents with Verbal Reinforcement Learning.arXiv preprint arXiv:2303.11366, 2023. doi: 10.48550/arXiv.2303.11366. URL https://arxiv.org/abs/ 2303.11366 on PithPith's reviewarXivdoi:10.48550/arxiv.2303.11366
  16. [17] David Silver and Richard S. Sutton. Welcome to the Era of Experience. Preprint of a chapter forDesigning an Intelligence, MIT Press, 2025. URL https://storage.googleapis.com/ deepmind-media/Era-of-Experience%20/The%20Era%20of%20Experience%20Paper. pdf on Pith
  17. [18] Xingyao Wang, Yangyi Chen, Lifan Yuan, Yizhe Zhang, Yunzhu Li, Hao Peng, and Heng Ji. Executable Code Actions Elicit Better LLM Agents.arXiv preprint arXiv:2402.01030, 2024. doi: 10.48550/arXiv. 2402.01030. URLhttps://arxiv.org/abs/2402.01030 on PithPith's reviewarXivdoi:10.48550/arxiv.2402.01030
  18. [19] Yinjie Wang, Xuyang Chen, Xiaolong Jin, Mengdi Wang, and Ling Yang. OpenClaw-RL: Train Any Agent Simply by Talking.arXiv preprint arXiv:2603.10165, 2026. doi: 10.48550/arXiv.2603.10165. URLhttps://arxiv.org/abs/2603.10165 on PithPith's reviewarXivdoi:10.48550/arxiv.2603.10165
  19. [20] Yinjie Wang, Tianbao Xie, Ke Shen, Mengdi Wang, and Ling Yang. RLAnything: Forge Environment, Policy, and Reward Model in Completely Dynamic RL System.arXiv preprint arXiv:2602.02488, 2026. doi: 10.48550/arXiv.2602.02488. URLhttps://arxiv.org/abs/2602.02488 on Pithdoi:10.48550/arxiv.2602.02488
  20. [21] Yang Xiao, Yusong Sun, Haoyi Wu, Wenyang Hui, Wen Da, Zhaokai Luo, Mu Chuan, Yao Hu, Wenjie Li, and Chengyue Jiang. PILOT in the Loop: Live Self-Improvement for Long-Horizon Agents.arXiv preprint arXiv:2608.26530, 2026. doi: 10.48550/arXiv.2608.26530. URL https: //arxiv.org/ab... on PithPith's reviewarXivdoi:10.48550/arxiv.2608.26530
show 9 more
  1. [22] Yifan Yang, Ziyang Gong, Weiquan Huang, Qihao Yang, Ziwei Zhou, Zisu Huang, Yan Li, Xuemei Gao, Qi Dai, Bei Liu, Kai Qiu, Yuqing Yang, Dongdong Chen, Xue-Ting Yang, and Chong Luo. SkillOpt: Executive Strategy for Self-Evolving Agent Skills.arXiv preprint arXiv:2605.23904, 2026... on PithPith's reviewarXiv
  2. [23] Shunyu Yao, Jeffrey Zhao, Dian Yu, Nan Du, Izhak Shafran, Karthik Narasimhan, and Yuan Cao. ReAct: Synergizing Reasoning and Acting in Language Models. InInternational Conference on Learning Representations, 2023. URLhttps://arxiv.org/abs/2210.03629. 19 ScienceBuddy: Recursive... on PithPith's reviewarXiv
  3. [24] Ya-Qi Yu, Hao Wang, Fangyu Hong, Xiangyang Qu, Gaojie Wu, Qiaoyu Luo, Nuo Xu, Huixin Wang, Wuheng Xu, Yongxin Liao, Zihao Chen, Haonan Li, Ziming Li, Dezhi Peng, Minghui Liao, Jihao Wu, Haoyu Ren, and Dandan Tu. Reinforcement Learning with Robust Rubric Rewards.arXiv preprint ... on PithPith's reviewarXiv
  4. [25] Jenny Zhang, Shengran Hu, Cong Lu, Robert Lange, and Jeff Clune. Darwin Godel Machine: Open-Ended Evolution of Self-Improving Agents.arXiv preprint arXiv:2505.22954, 2025. doi: 10.48550/arXiv.2505.22954. URLhttps://arxiv.org/abs/2505.22954 on PithPith's reviewarXivdoi:10.48550/arxiv.2505.22954
  5. [26] Jenny Zhang, Bingchen Zhao, Wannan Yang, Jakob Foerster, Jeff Clune, Minqi Jiang, Sam Devlin, and Tatiana Shavrina. Hyperagents.arXiv preprint arXiv:2603.19461, 2026. doi: 10.48550/arXiv. 2603.19461. URLhttps://arxiv.org/abs/2603.19461 on PithPith's reviewdoi:10.48550/arxiv.2603.19461
  6. [27] Qizheng Zhang, Changran Hu, Shubhangi Upasani, Boyuan Ma, Fenglu Hong, Vamsidhar Kamanuru, Jay Rainton, Chen Wu, Mengmeng Ji, Hanchen Li, Urmish Thakker, James Zou, and Kunle Olukotun. Agentic Context Engineering: Evolving Contexts for Self-Improving Language Models.arXiv prep... on PithPith's reviewarXivdoi:10.48550/arxiv.2510.04618
  7. [28] Adam Zweiger, Jyothish Pari, Han Guo, Ekin Akyürek, Yoon Kim, and Pulkit Agrawal. Self-Adapting Language Models.arXiv preprint arXiv:2506.10943, 2025. doi: 10.48550/arXiv.2506.10943. URL https://arxiv.org/abs/2506.10943. 20 ScienceBuddy: Recursive-in-Recursive Self-Improvement... on Pithdoi:10.48550/arxiv.2506.10943
  8. [2024] doi: 10.48550/arXiv.2407.10362. URLhttps://arxiv.org/abs/2407.10362 on PithPith's reviewarXivdoi:10.48550/arxiv.2407.10362
  9. [2026] URLhttps://arxiv.org/abs/2606.12834 on PithPith's reviewarXiv
Cited by 9 Pith papers

Challenges and discussion

Disagree with the review, or know something it missed? Leave a comment, or challenge this review formally: a challenge becomes a permanent entry in this paper's record, and its decision is published here.

No comments yet.

Sign in to comment