Tag
This paper introduces a controlled evaluation framework that separates answer coverage from repeatable task specialization in LLM-generated harnesses on MATH-500, finding that additional coverage largely comes from repeated executions of the same program rather than genuine specialization gains.
The paper introduces Task Specialization Fine-Tuning (TSFT) for contextual reinforcement learning, a framework that uses a simple parametric model and integer linear programming to allocate fine-tuning budgets efficiently, significantly outperforming baselines in task coverage.
This paper reviews eight years of language model evolution from BERT to advanced agentic systems, documenting the collapse of the capability-cost curve and the emergence of task-targeted models through empirical benchmarks and reproducible experiments.