Related papers: SFT-GRPO Data Overlap as a Post-Training Hyperpara…
Code-agent RL often receives weak feedback: rollout-time signals are reliable and executable, but capture only necessary or surface conditions for task success rather than the target semantic predicate. Using agentic compile-fix as the…
Mathematical reasoning is a key benchmark for large language models. Reinforcement learning is a standard post-training mechanism for improving the reasoning capabilities of large language models, yet performance remains sensitive to the…
Large language models (LLMs) can now solve complex problems through long chain-of-thought (CoT) reasoning, but the trade-off between performance and token cost remains a central challenge. To address this issue, supervised fine-tuning (SFT)…
Task-adaptive pre-training (TAPT) and Self-training (ST) have emerged as the major semi-supervised approaches to improve natural language understanding (NLU) tasks with massive amount of unlabeled data. However, it's unclear whether they…
Large reasoning models that use long chain-of-thought excel at problem-solving yet waste compute on redundant checks. Curbing this overthinking is hard: training-time length penalties can cripple ability, while inference-time early-exit…
Reinforcement learning (RL) has proven effective in incentivizing the reasoning abilities of large language models (LLMs), but suffers from severe efficiency challenges due to its trial-and-error nature. While the common practice employs…
The pre-training and fine-tuning paradigm has contributed to a number of breakthroughs in Natural Language Processing (NLP). Instead of directly training on a downstream task, language models are first pre-trained on large datasets with…
This paper investigates various approaches using Large Language Models (LLMs) to identify gaps and misconceptions in students' self-explanations of specific instructional material, in our case explanations of code examples. This research is…
Continual post-training of generative models is widely used, yet a principled understanding of when and why forgetting occurs remains limited. We develop theoretical results under a two-mode mixture abstraction (representing old and new…
Supervised fine-tuning (SFT) is the standard approach for binary classification tasks such as toxicity detection, factuality verification, and causal inference. However, SFT often performs poorly in real-world settings with label noise,…
We introduce a novel method that averages the logits of a frozen reference policy (e.g., SFT) and a trainable policy, and incorporate the method into Group Relative Policy Optimization (GRPO). In contrast to Reinforcement Learning with…
Generative speech enhancement offers a promising alternative to traditional discriminative methods by modeling the distribution of clean speech conditioned on noisy inputs. Post-training alignment via reinforcement learning (RL) effectively…
Simulation offers a scalable and low-cost way to enrich vision-language-action (VLA) training, reducing reliance on expensive real-robot demonstrations. However, most sim-real co-training methods rely on supervised fine-tuning (SFT), which…
This paper delves into the continuous post-training optimization methods for small language models, and proposes a continuous post-training alignment data construction method for small language models. The core of this method is based on…
Graph-matching metrics such as Smatch are the de facto standard for evaluating neural semantic parsers, yet they capture surface overlap rather than logical equivalence. We reassess evaluation by pairing graph-matching with automated…
Modern large language models (LLMs) are increasingly fine-tuned via reinforcement learning from human feedback (RLHF) or related reward optimisation schemes. While such procedures improve perceived helpfulness, we investigate whether…
Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL) are two prominent post-training paradigms for refining the capabilities and aligning the behavior of Large Language Models (LLMs). Existing approaches that integrate SFT and RL…
Recent mixed-policy optimization methods for LLM reasoning that interleave or blend supervised and reinforcement learning signals report improvements over the standard SFT-then-RL pipeline. We show that numerous recently published research…
Language model (LM) post-training relies on two stages of human supervision: task demonstrations for supervised finetuning (SFT), followed by preference comparisons for reinforcement learning from human feedback (RLHF). As LMs become more…
We present a two-stage fine-tuning approach to make the large language model Qwen3 14B "think" natively in Korean. In the first stage, supervised fine-tuning (SFT) on a high-quality Korean reasoning dataset establishes a strong foundation…