arXiv:2610. 33882v1 Announce Type: cross Abstract: Hybrid post-training often mixes supervised fine-tuning (SFT) and reinforcement learning (RL), using training-side signals to set each sample's balance.
arXiv:2604. 10688v2 Announce Type: replace-cross Abstract: On-policy reinforcement learning has become the dominant paradigm for reasoning alignment in large language models, yet its sparse, outcome-level rewards make token-level credit assignment notoriously difficult.
arXiv:2609.06107v1 Announce Type: new Abstract: Data policies for reinforcement learning with verifiable rewards (RLVR) determine which rollouts are used, how strongly they are weighted, and which do...
LLM-as-judge is essential for evaluating open-ended text and steering post-training, yet improving the judge itself typically relies on expensive annotations, reward models, or distillation from stron...
arXiv:2606. 12634v1 Announce Type: cross Abstract: Long-horizon tool-use reinforcement learning can learn from outcome verification, but its trajectory-level advantage is broadcast across many reasoning, API, and answer tokens.
Steering vectors (SVs) are residual-stream directions that can shift LLM behavior when added to hidden states during inference. This study investigates whether the same SVs can guide parameter updates during post‑training, exploring supervised learning (SFT and distillation) and reinforcement learning with verifiable rewards (RLVR). The authors find that SV injection improves performance over plain SFT and distillation, achieving up to a 12% gain in certain distillation settings, and that applying a correctness‑contrast direction as advantage repair in RL recovers learning signals for held‑out math benchmarks. They also identify failure boundaries related to code generation, cross‑family layer transfer, and inference‑time application on SV‑trained checkpoints.
arXiv:2606. 05718v1 Announce Type: cross Abstract: On-policy distillation (OPD) improves reasoning by training a student on trajectories sampled from its own policy under supervision from a teacher.
The paper introduces Cross-Behavior Policy Optimization (CBPO), a method that unifies offline demonstrations and policy-generated rollouts into a single regression loss using a fixed-reference target. CBPO eliminates the need for separate supervised fine‑tuning and reinforcement learning objectives, allowing positive and negative offline completions, reference responses, and current and historical rollouts to share a common reward‑labeled format and replay buffer. Experiments on two model scales and four mathematical reasoning benchmarks show that CBPO, especially the CBPO‑Unified variant, outperforms baseline methods in avg@32 and pass@32 metrics, with historical rollouts providing additional gains.
arXiv:2605.28791v2 Announce Type: replace-cross Abstract: On-policy self-distillation (SD) improves LLM reasoning by using teacher-side privileged information (PI) to turn sparse verifier outcomes in...
arXiv:2607. 04412v1 Announce Type: new Abstract: Reinforcement learning (RL) for non-verifiable instruction following increasingly relies on LLM judges with prompt-specific rubrics as reward signals.
arXiv:2609.36546v1 Announce Type: cross Abstract: On-policy distillation (OPD) trains a student model on its self-generated trajectories with dense token-level teacher feedback. However, naive OPD ma...
arXiv:2606. 29280v1 Announce Type: cross Abstract: We identify intervention bias as a previously unquantified failure mode of zero-shot large-language-model (LLM) educational advisory agents: without task-specific training, they recommend action when a hindsight-optimal oracle policy mandates inaction.
The paper introduces a recursive self-improvement framework for language models that replaces an external teacher with a frozen copy of the student, enabling dynamic co-evolution (DCE) and self-refined concise learning (SRCL). DCE allows the privileged teacher to evolve alongside the student, while SRCL trains on shorter, verified rewrites to reduce verbosity. Experiments show that the combined DCE+SRCL approach outperforms traditional on‑policy self‑distillation across multiple model sizes and math benchmarks, achieving significant accuracy gains and shorter outputs.