AllenAI OLMo 3 技术报告深度解析
报告链接：https://www.datocms-assets.com/64837/1763662397-1763646865-olmo_3_techn
...
HuggingFace 高分论文：首个达到 IPhO 金牌水平的开源模型是如何炼成的？
论文标题：P1: Mastering Physics Olympiads with Reinforcement Learning
论文链接：http
...
Meta 提出 SoCE 策略，仅靠权重融合实现 SOTA
论文标题：Souper-Model: How Simple Arithmetic Unlocks State-of-the-Art LLM Perfor
...
陈丹琦团队新作 Retaining by Doing：揭示 RL 比 SFT 为什么更能缓解灾难性遗忘
论文标题：Retaining by Doing: The Role of On-Policy Data in Mitigating Forgetting
...
微软提出 GAD：通过生成对抗蒸馏方法实现 On-Policy 蒸馏 GPT-5
论文标题：Black-Box On-Policy Distillation of Large Language Models
论文链接：https:
...
LightReasoner：利用小模型引导大模型推理的对比学习框架
论文标题：LIGHTREASONER: CAN SMALL LANGUAGE MODELS TEACH LARGE LANGUAGE MODELS RE
...
WeiBo AI 推出 1.5B 小模型，成本实现 SOTA 级推理
论文标题：Tiny Model, Big Logic: Diversity-Driven Optimization Elicits Large-Mode
...
Meta FAIR 推出 HERO：LLM 强化中集成稀疏与密集奖励
论文标题：Hybrid Reinforcement: When Reward Is Sparse, It’s Better to Be Dense
...
专挑模型的“软肋”下手：阿里 MIWV 如何实现用1%数据超越全量微调？
论文标题：Importance-Aware Data Selection for Efficient LLM Instruction Tuning
...
Meta AI 推出 RIFL：基于准则的强化学习来提升 LLM 指令遵循能力
论文标题：Rubric-Based Benchmarking and Reinforcement Learning for Advancing LLM
...