ImproveAnyTask: An Autonomous Post-Training Harness for Iterative Model Self-Improvement
cs.AI
Submitted: 2026-10-05
Updated: 2026-10-05
Comments: 18 pages, 4 figures
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- OpenCodeReasoning: Advancing Data Distillation for Competitive Coding
- Pioneer Agent: Continual Improvement of Small Language Models in Production
- LLMZero: Discovering Adaptive Training Strategies for RL Post-Training via LLM Agents
- GLM-5: from Vibe Coding to Agentic Engineering
- Linguistic and Argument Diversity in Synthetic Data for Function-Calling Agents
- OpenThoughts: Data Recipes for Reasoning Models
- DeReason: A Difficulty-Aware Curriculum Improves Decoupled SFT-then-RL Training for General Reasoning
- Tailoring the Curriculum: Student-Centered Reasoning Distillation via Dynamic Data-Model Compatibility
- LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code
- Kimi K3: Open Frontier Intelligence
- FT-Dojo: Towards Autonomous LLM Fine-Tuning with Language Agents
- Let's Verify Step by Step
- What is Missing from AI Post-Training AI: An Empirical Analysis
- SuperGPQA: Scaling LLM Evaluation across 285 Graduate Disciplines
- General-Reasoner: Advancing LLM Reasoning Across All Domains
- TREX: Automating LLM Fine-tuning via Agent-Driven Tree-based Exploration
- MLE-STAR: Machine Learning Engineering Agent via Search and Targeted Refinement
- Generalizing Verifiable Instruction Following
- PostTrainBench: Can LLM Agents Automate LLM Post-Training?
- GPQA: A Graduate-Level Google-Proof Q&A Benchmark
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems