cs.CL（2026-02-15）

📊 共 12 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (9 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (3)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (9 篇)

#题目一句话要点标签🔗⭐
1 Chain-of-Thought Reasoning with Large Language Models for Clinical Alzheimer's Disease Assessment and Diagnosis 利用大型语言模型和思维链推理进行临床阿尔茨海默病评估与诊断 large language model chain-of-thought
2 LogitsCoder: Towards Efficient Chain-of-Thought Path Search via Logits Preference Decoding for Code Generation LogitsCoder：通过Logits偏好解码实现高效的代码生成思维链路径搜索 chain-of-thought
3 Investigation for Relative Voice Impression Estimation 提出相对语音印象估计框架，利用自监督语音表征捕捉细微感知差异。 large language model multimodal
4 The Sufficiency-Conciseness Trade-off in LLM Self-Explanation from an Information Bottleneck Perspective 从信息瓶颈视角研究LLM自解释的充分性-简洁性权衡 large language model chain-of-thought
5 Knowing When Not to Answer: Abstention-Aware Scientific Reasoning 提出一种基于拒绝回答的科学推理框架，提升科学结论的可靠性。 large language model ✅
6 GPT-5 vs Other LLMs in Long Short-Context Performance 评估GPT-5等LLM在长短上下文社交媒体抑郁检测中的性能退化 large language model
7 Annotation-Efficient Vision-Language Model Adaptation to the Polish Language Using the LLaVA Framework 利用LLaVA框架，通过高效标注方法将视觉-语言模型适配到波兰语 multimodal
8 CCiV: A Benchmark for Structure, Rhythm and Quality in LLM-Generated Chinese \textit{Ci} Poetry CCiV：构建中文词牌诗生成基准，评估LLM在结构、韵律和质量上的表现 large language model
9 HLE-Verified: A Systematic Verification and Structured Revision of Humanity's Last Exam 提出HLE-Verified，通过系统验证和修订提升HLE基准的可靠性。 large language model ✅

🔬 支柱二：RL算法与架构 (RL & Architecture) (3 篇)

#题目一句话要点标签🔗⭐
10 Detecting LLM Hallucinations via Embedding Cluster Geometry: A Three-Type Taxonomy with Measurable Signatures 通过嵌入聚类几何结构检测LLM幻觉，提出可测量特征的三种类型分类法 distillation large language model
11 The Interspeech 2026 Audio Reasoning Challenge: Evaluating Reasoning Process Quality for Audio Reasoning Models and Agents Interspeech 2026音频推理挑战赛：评估音频推理模型和智能体的推理过程质量 reinforcement learning chain-of-thought
12 Open Rubric System: Scaling Reinforcement Learning with Pairwise Adaptive Rubric 提出OpenRS，通过可检验的原则和自适应评分标准提升开放场景强化学习的鲁棒对齐。 reinforcement learning

⬅️ 返回 cs.CL 首页 · 🏠 返回主页