Rethinking the Evaluation of Harness Evolution for Agents

• URL: http://arxiv.org/abs/2607.12227v1
• Date: Tue, 14 Jul 2026 00:18:42 GMT
• Title: Rethinking the Evaluation of Harness Evolution for Agents
• Abstract summary: Existing harness evolution methods use unit test cases to search for harness configurations and then report final performance on the same public benchmark.<n>This protocol raises two fundamental concerns.<n>As in agentic test-time scaling, it should therefore be compared with simple task-level search baselines under matched feedback and inference budgets.
• Score: 83.07258924910069
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We revisit the evaluation of automatic harness evolution for LLM agents. Existing harness evolution methods use unit test cases to search for harness configurations and then report final performance on the same public benchmark. This protocol raises two fundamental concerns. First, harness evolution is itself an iterative search procedure that repeatedly evaluates and revises candidate harnesses using task feedback. As in agentic test-time scaling, it should therefore be compared with simple task-level search baselines under matched feedback and inference budgets to determine whether its gains arise from improved harness design or from additional search alone. Second, because the search and the final evaluation share the same benchmark, the reported gains risk overfitting to that specific task set. To address these concerns, we conduct an extensive evaluation comparing harness evolution with simple test-time scaling and discovery baselines under comparable feedback and inference budgets, and also evaluate evolved harnesses on held-out tasks to assess whether the discovered improvements generalize. Experiments on Terminal-Bench 2.1 with GPT-5.4 and Claude Opus 4.6 show that automatic harness evolution does not consistently outperform simple test-time scaling methods and exhibits limited generalization. Our results raise important questions about the effectiveness of automatic harness evolution and highlight the need for fairer evaluation protocols and benchmarks for automatic harness design. Our code is available at https://github.com/rethinking-harness-evolution.

Related papers

• HarnessDev: Can LLMs Create and Evolve Their Own Agent Harness? [61.02699867449589]
  We introduce HarnessDev, a benchmark that shifts the unit of evaluation from task outputs to run infrastructure.<n>In Creation, the agent starts from a minimal seed and a small number of cases, then builds a complete execution system.<n>In Evolution, it starts from its own created harness and iteratively revises it using downstream execution feedback, with the goal of improving benchmark performance.
  arXiv  Detail & Related papers  (2026-09-01T15:45:33Z)
• Verify Smarter, Evolve Further: Efficient Harness Evolution through Behavior-Aware Verification [30.941867057615223]
  Existing propose-and-verify methods score every candidate on a fixed task set.<n>We introduce HarnessLens, a budget-aware framework for automated harness evolution.
  arXiv  Detail & Related papers  (2026-08-27T16:12:23Z)
• Task-CoEvolve: Efficient Harness Optimization via Adaptive Validation Task Selection [51.39973047614183]
  We present a novel approach to efficient LLM harness optimization through adaptive validation task selection.<n>$textbfTask-CoEvolve$, which co-evolves the validation tasks with the harness by addressing two challenges: selecting informative tasks and estimating full-set performance from partial evaluations.
  arXiv  Detail & Related papers  (2026-08-20T15:24:54Z)
• Automated Discovery Has No Universally Superior Harness [63.53666712024897]
  We decompose OpenEvolve-style evolutionary search and the TTT-Discover search harness into its constituent components.<n>We evaluate 30 budget-matched harnesses across 12 model-problem pairs.
  arXiv  Detail & Related papers  (2026-07-20T17:59:37Z)
• TTHE: Test-Time Harness Evolution [50.26245555541721]
  Existing approaches optimize such harnesses before deployment, searching training or development data for a fixed agent workflow that is then frozen at test time.<n>We ask whether the harness can instead be optimized during evaluation itself, using only the unlabeled execution traces the agent produces on the test inputs.<n>During evaluation, TTHE maintains a population of candidate harnesses and refines them through an agentic proposer that reasons over their execution traces.<n>A judge then commits an improved harness from execution-derived proxy signals, and the selected program persists to govern subsequent inputs.
  arXiv  Detail & Related papers  (2026-07-09T05:53:39Z)
• BenchTrace: A Benchmark for Testing Reflection Ability and Controlled Evolution in LLM Agents [24.997265534888626]
  We present textbfBenchTrace, a benchmark for evaluating self-evolution ability in LLM agents.<n>BenchTrace is built on a snapshot-reflection dataset of 1,821 annotated episodes spanning six diverse tasks.<n>We propose textbffailure avoidance rate (FAR), a new evaluation metric measuring the fraction of test cases in which the agent successfully avoids the target failure instance.
  arXiv  Detail & Related papers  (2026-05-28T01:25:37Z)
• What Do Evolutionary Coding Agents Evolve? [46.561777689365165]
  EvoTrace is a dataset of evolutionary coding traces spanning four evolutionary frameworks, reasoning and non-reasoning models, and 16 tasks across mathematics and algorithm design.<n>We analyze these traces with EvoReplay, a replay-based methodology that reconstructs the local search states behind high-scoring solutions.<n>Across EvoTrace, most score gains come from a small subset of these edit types.
  arXiv  Detail & Related papers  (2026-05-19T16:41:45Z)
• FRESCO: Benchmarking and Optimizing Re-rankers for Evolving Semantic Conflict in Retrieval-Augmented Generation [73.22935457705057]
  We introduce FRESCO, a benchmark for evaluating re-rankers in temporally dynamic contexts.<n>By pairing recency-seeking queries with historical Wikipedia revisions, FRESCO tests whether re-rankers can prioritize factually recent evidence while maintaining semantic relevance.<n>Our evaluation reveals a consistent failure mode across existing re-rankers: a strong bias toward older, semantically rich documents, even when they are factually obsolete.
  arXiv  Detail & Related papers  (2026-04-14T17:04:25Z)
• Towards Better Evolution Modeling for Temporal Knowledge Graphs [0.0]
  Temporal knowledge graphs (TKGs) preserve evolving human knowledge.<n>Recent research has focused on designing models to learn the evolutionary nature of TKGs to predict future facts.<n>We find that existing benchmarks inadvertently introduce a shortcut.
  arXiv  Detail & Related papers  (2026-02-09T07:37:40Z)
• Chain-of-Retrieval Augmented Generation [91.02950964802454]
  This paper introduces an approach for training o1-like RAG models that retrieve and reason over relevant information step by step before generating the final answer.<n>Our proposed method, CoRAG, allows the model to dynamically reformulate the query based on the evolving state.
  arXiv  Detail & Related papers  (2025-01-24T09:12:52Z)
• Solving Offline Reinforcement Learning with Decision Tree Regression [0.0]
  This study presents a novel approach to addressing offline reinforcement learning problems by reframing them as regression tasks. We introduce two distinct frameworks: return-conditioned and return-weighted decision tree policies. Despite the simplification inherent in this reformulated approach to offline RL, our agents demonstrate performance that is at least on par with the established methods.
  arXiv  Detail & Related papers  (2024-01-21T23:50:46Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.