Learning to Evaluate Before Improving: Automatic Rubric Induction for Automatic Research Agents

Hugging Face Daily Papers Papers

Summary

AutoSciRub is an evaluation-first framework that improves autonomous scientific agents by generating task-specific executable rubrics to guide experiments and verification, achieving consistent performance gains on benchmarks.

Autonomous scientific research agents are increasingly applied to end-to-end scientific workflows, including literature review, data analysis, experimentation, and report generation. However, open-ended research tasks often do not clearly specify the analyses, methods, and success criteria required to complete the task. As a result, agents may miss important analyses, use inappropriate methods, or draw conclusions that are insufficiently supported by evidence. To address the problem, we present AutoSciRub, an evaluation-first framework that induces a task-specific executable rubric before research execution, and uses it to guide execution, criterion-level verification as well as iterative revision. AutoSciRub decomposes an underspecified instruction into atomic scientific goals, grounds them in relevant literature and task-visible data, and synthesizes specific, actionable, and verifiable criteria. The resulting rubric makes implicit experimental and evidential requirements explicit, providing guidance for experiments and analyses. During revision, rubric-guided verification identifies unmet criteria and enables targeted refinement of the research report and its supporting artifacts. On ResearchClawBench, AutoSciRub consistently improves all tested configurations, with an average gain of 2.08 points across three backbone LLMs under the fixed Codex harness and 2.95 points across three agent harnesses using a fixed DeepSeek-V4-Flash backbone. On a randomly sampled 20-task subset of AstaBench E2E Discovery, AutoSciRub further achieves an average improvement of 16.8 points across three agent harnesses, while maintaining or increasing the number of successfully completed tasks. These results demonstrate that evaluation-first guidance provides an effective and generalizable control mechanism for autonomous scientific research (Code: https://github.com/zjunlp/AutoSciRub).
Original Article
View Cached Full Text

Cached at: 09/01/26, 11:53 AM

Paper page - Learning to Evaluate Before Improving: Automatic Rubric Induction for Automatic Research Agents

Source: https://huggingface.co/papers/2608.31076

Abstract

AutoSciRub improves autonomous scientific agents by generating task-specific executable rubrics that guide experiments, verify criteria, and iteratively refine outputs.

Autonomous scientific research agents are increasingly applied to end-to-end scientific workflows, including literature review, data analysis, experimentation, and report generation. However, open-ended research tasks often do not clearly specify the analyses, methods, and success criteria required to complete the task. As a result, agents may miss important analyses, use inappropriate methods, or draw conclusions that are insufficiently supported by evidence. To address the problem, we presentAutoSciRub, an evaluation-first framework that induces a task-specificexecutable rubricbefore research execution, and uses it to guide execution, criterion-level verification as well as iterative revision.AutoSciRubdecomposes an underspecified instruction intoatomic scientific goals, grounds them in relevant literature and task-visible data, and synthesizes specific, actionable, and verifiable criteria. The resulting rubric makes implicit experimental and evidential requirements explicit, providing guidance for experiments and analyses. During revision,rubric-guided verificationidentifies unmet criteria and enables targeted refinement of the research report and its supporting artifacts. OnResearchClawBench,AutoSciRubconsistently improves all tested configurations, with an average gain of 2.08 points across three backbone LLMs under the fixed Codex harness and 2.95 points across three agent harnesses using a fixed DeepSeek-V4-Flash backbone. On a randomly sampled 20-task subset ofAstaBenchE2E Discovery,AutoSciRubfurther achieves an average improvement of 16.8 points across three agent harnesses, while maintaining or increasing the number of successfully completed tasks. These results demonstrate that evaluation-first guidance provides an effective and generalizable control mechanism for autonomous scientific research (Code: https://github.com/zjunlp/AutoSciRub).

View arXiv pageView PDFGitHub14Add to collection

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2608.31076 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2608.31076 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2608.31076 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

An Empirical Study of Automating Agent Evaluation

arXiv cs.CL

This paper introduces EvalAgent, a system that automates the evaluation of AI agents by encoding domain-specific expertise, addressing the limitations of standard coding assistants in this task. It also presents AgentEvalBench, a benchmark for testing evaluation pipelines, and demonstrates significant improvements in evaluation reliability.

RUBAS: Rubric-Based Reinforcement Learning for Agent Safety

arXiv cs.LG

RUBAS is a rubric-based reinforcement learning framework for agent safety that decomposes LLM agent behavior into four dimensions—tool-use safety, argument safety, response safety, and helpfulness—providing fine-grained rewards over complete trajectories. Experiments show RUBAS improves safety over standard alignment baselines while reducing tool-grounded hallucinations and maintaining competitive utility.