This work asks whether replacing free-form reasoning with a restricted, executable program turns this sparse reward into a graded, more interpretable training signal, using a standard SFT → RLOO pipeline as the reference point.
This work shows that training a strong instruction-tuned reasoning model on its own answer-conditioned chains sharply lowers its verifiable-reasoning accuracy, and generates answer-blind data, because no correctness filter can see this damage in the data.
Jungseob Lee, Seungyoon Lee, Suhyune Son et al.· 0 citations
Transparent educational question answering asks for answers that are not only correct but explainable, and doing so with small models rules out the reasoning power of the largest proprietary systems. The EXACT 2026 competition poses this problem concretely: open-weight language models of at most 8B parameters, self-hosted, with a natural-language explanation for every answer. It pairs two tasks: logical reasoning over university regulations, and multi-step physics problem solving. We describe the system that team \cotu{} developed to address both, a neuro-symbolic Program-of-Thought pipeline in which a 4B backbone writes a program rather than stating an answer directly: for regulation queries it emits a Z3 encoding whose entailment verdict grounds the deduction, and for physics it emits numerical Python, both wrapped in a shared self-correction loop and a unified explained-JSON output. Answer-type routing, distillation-based task fine-tuning, and a latency-aware serving stack -- SGLang with speculative decoding -- keep the system within the 60-second per-query limit. The system achieved a \textbf{perfect score} on the physics task in both automated selection rounds and obtained the \textbf{highest final-round technical score} of any team -- $13.44/15$, combining automated answer evaluation with expert-judged reasoning depth -- with the equally weighted presentation score included, \cotu{} placed 3rd overall. Grounding answers in a symbolic solver yields correct, verifiable deductions at the 4B scale, and the residual difficulty lies in premise selection rather than the deduction itself.
Quoc-Khang Tran, M. Nguyen, P. Thai et al.· 0 citations
This work systematize the RL-for-LLM paradigm and provides a compute-centric analysis of prominent post-training algorithmic frameworks: Proximal Policy Optimization (PPO), Group Relative Policy Optimization (GRPO), as well as their variants, and develops a taxonomy of intra- and inter-model parallelism strategies for RL-for-LLMs.
Maciej Besta, Leonard Schmidt, Lara Nonino et al.· 0 citations
Reason Popper-ly, a neurosymbolic framework that uses inductive logic programming (ILP) to learn relation composition rules from reasoning traces and deploys them as an online verifier for step-level correction, consistently improves terminal accuracy over standard CoT.
A unified supervision framework is introduced that embeds programmatically verifiable checkers into synthesized instruction-conflict instances, enabling alignment without oracle labels or reasoning traces, supporting both instruction-tuned and reasoning models.
Sian-Yao Huang, Li-Hsien Chang, Che-Yu Lin et al.· Neural Information Processin...· 4 citations
This work makes execution time learnable through three stages: how code is tested, by building DMC-Optim with large optimization tests and a calibrated sandbox; how speed is turned into reward, by composing correctness and speed in the RL environment and using an offline simulator to predict the most promising configurations.
Pierre Chambon, Kunhao Zheng, Juliette Decugis et al.· 0 citations