CalibForge: Adversarial Solver Calibration for Scaling Learnable Terminal Tasks

• URL: http://arxiv.org/abs/2608.06352v1
• Date: Thu, 06 Aug 2026 17:53:18 GMT
• Title: CalibForge: Adversarial Solver Calibration for Scaling Learnable Terminal Tasks
• Abstract summary: CalibForge is an autonomous terminal-task synthesis system.<n>It uses verified solver behavior to revise candidate tasks through adversarial solver calibration.<n>Using CalibForge, we construct 5,431 terminal tasks.
• Score: 87.1086886770395
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Training terminal agents requires executable and verifiable tasks that are not merely solvable, but appropriately challenging for learning. Executable validation establishes feasibility, yet does not reveal how a task behaves relative to a given solver setting. In this paper, we present CalibForge, an autonomous terminal-task synthesis system that uses verified solver behavior to revise candidate tasks through adversarial solver calibration. Multi-solver calibration targets disagreement within a heterogeneous solver pool, whereas contrastive solver calibration targets a designated strong-pass/weak-fail relation; both operationalize a solver-relative learnable zone anchored in demonstrated solvability. Using CalibForge, we construct 5,431 calibrated terminal tasks. Our ablations show that both strategies yield more effective supervision than authoring and validation alone or ordinary single-solver feedback. Models trained on the full collection achieve 32.58% and 47.57% on Terminal-Bench 2.0. The largest improvements over the corresponding base model reach 24.71 percentage points on Terminal-Bench 2.0, 27.68 points on SWE-bench Pro, and 30.04 points on Doc2Repo. Together, these results support solver-relative learnability as a practical target for constructing effective and transferable agent training data.

Related papers

• TRACE: A Self-Evolving Skill Bank for Consistent, Limit-Aware LLM Agents [41.82249291285459]
  TRACE iteratively improves a skill-based agent's behavioral knowledge without modifying model weights.<n>On GPT-5.5, TRACE improves consistency (Pass3) by 34.6 points, from 59.9% to 94.5%, while shrinking the gap between potential and reliable performance to just 4.0 points.<n>These results show that TRACE converts high model potential into stable, consistent performance gain.
  arXiv  Detail & Related papers  (2026-08-24T04:36:03Z)
• ClawTrack: Towards Trace-Level Evaluation and Improvement of Real-World Autonomous Agents [23.840537431318296]
  We present ClawTrack, a dual-assessment benchmark that simultaneously measures what an agent achieves and how it achieves it.<n>ClawTrack comprises 320 tasks across 8 domains with 25+ deterministic mock services.<n>A Process Grader scores each reasoning turn along four dimensions (goal alignment, efficiency, information utilization, and result verification) anchored by 12,541 task-specific items.
  arXiv  Detail & Related papers  (2026-07-30T11:18:47Z)
• Efficient Visual Pointing for Embodied AI:Agent-Driven Data Synthesis, Cross-Block Attention, and Iterative Correction [55.11480729304395]
  We describe our PointArena 2026 solution, which achieves 77.2% overall accuracy and ranks second on the benchmark.<n>The ap proach targets three failure modes. First, agent-driven syn thesis builds large semantic and anchor-relative candidate pools.<n>Second, a determinis tic steerable-data pipeline creates a verified 10,000-sample main set, plus reserve samples, using masks, templates, and path verification.
  arXiv  Detail & Related papers  (2026-06-29T06:39:03Z)
• SENTINEL: Failure-Driven Reinforcement Learning for Training Tool-Using Language Model Agents [52.758563996979596]
  Language model agents are increasingly effective in solving realistic tasks through multi-turn tool use.<n>We propose SENTINEL, a failure-driven reinforcement learning framework that turns rollout failures into targeted training tasks.<n>On Tau2-Bench Retail with Qwen3-4B-Thinking-2507, SENTINEL improves Pass1 from 66.4 to 74.9 and outperforms RL on general synthetic tasks across Passk metrics.
  arXiv  Detail & Related papers  (2026-06-11T05:06:50Z)
• Socratic-SWE: Self-Evolving Coding Agents via Trace-Derived Agent Skills [12.442806027914097]
  We introduce Socratic-SWE, a closed-loop self-evolution framework that reuses the agent's historical solving traces as a source of training signal.<n>Rather than treating traces only as evidence for reward computation, Socratic-SWE distills them into structured agent skills that summarize recurring failures and effective repair patterns.
  arXiv  Detail & Related papers  (2026-06-05T16:00:17Z)
• MARGIN: Runtime Confidence Calibration for Multi-Agent Foundation Model Coordination [0.0]
  We present MARGIN (Multi-Agent Grading via Incremental Normalisation), an online calibration method that learns per-agent, per-band calibration factors from the task stream itself.<n>Across 18 foundation models, 8 benchmarks, and over 44,000 observations, MARGIN achieves 3-6x lower calibration error than the best design-time baseline under distribution shift.
  arXiv  Detail & Related papers  (2026-05-21T18:25:05Z)
• Eigen-1: Adaptive Multi-Agent Refinement with Monitor-Based RAG for Scientific Reasoning [53.45095336430027]
  We develop a unified framework that combines implicit retrieval and structured collaboration.<n>On Humanity's Last Exam (HLE) Bio/Chem Gold, our framework achieves 48.3% accuracy.<n>Results on SuperGPQA and TRQA confirm robustness across domains.
  arXiv  Detail & Related papers  (2025-09-25T14:05:55Z)
• Heimdall: test-time scaling on the generative verification [2.662648783972914]
  We propose Heimdall, the long CoT verification LLM that can accurately judge the correctness of solutions.<n>With pure reinforcement learning, we boost the verification accuracy from 62.5% to 94.5% on competitive math problems.<n>We also propose Pessimistic Verification to extend the functionality of Heimdall to scaling up the problem solving.
  arXiv  Detail & Related papers  (2025-04-14T15:46:33Z)
• SolBench: A Dataset and Benchmark for Evaluating Functional Correctness in Solidity Code Completion and Repair [51.0686873716938]
  We introduce SolBench, a benchmark for evaluating the functional correctness of Solidity smart contracts generated by code completion models.<n>We propose a Retrieval-Augmented Code Repair framework to verify functional correctness of smart contracts.<n>Results show that code repair and retrieval techniques effectively enhance the correctness of smart contract completion while reducing computational costs.
  arXiv  Detail & Related papers  (2025-03-03T01:55:20Z)
• Improving LLM Reasoning through Scaling Inference Computation with Collaborative Verification [52.095460362197336]
  Large language models (LLMs) struggle with consistent and accurate reasoning. LLMs are trained primarily on correct solutions, reducing their ability to detect and learn from errors. We propose a novel collaborative method integrating Chain-of-Thought (CoT) and Program-of-Thought (PoT) solutions for verification.
  arXiv  Detail & Related papers  (2024-10-05T05:21:48Z)
• V-STaR: Training Verifiers for Self-Taught Reasoners [71.53113558733227]
  V-STaR trains a verifier using DPO that judges correctness of model-generated solutions. Running V-STaR for multiple iterations results in progressively better reasoners and verifiers.
  arXiv  Detail & Related papers  (2024-02-09T15:02:56Z)

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.