cs.CL（2025-04-22）

📊 共 20 篇论文 | 🔗 4 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (13 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (7 🔗2)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (13 篇)

#题目一句话要点标签🔗⭐
1 Exploring Cognitive and Aesthetic Causality for Multimodal Aspect-Based Sentiment Analysis 提出Chimera框架以解决多模态情感分析中的认知与美学因果问题 large language model multimodal ✅
2 The Paradox of Poetic Intent in Back-Translation: Evaluating the Quality of Large Language Models in Chinese Translation 提出BT-Fried评估体系，揭示大语言模型汉英翻译中诗意理解的悖论。 large language model
3 PHYBench: Holistic Evaluation of Physical Perception and Reasoning in Large Language Models PHYBench：一个用于全面评估大语言模型物理感知与推理能力的新基准 large language model
4 A closer look at how large language models trust humans: patterns and biases 研究大型语言模型对人类的信任模式与偏差，揭示其决策过程中的潜在风险。 large language model
5 Automated Creativity Evaluation for Large Language Models: A Reference-Based Approach 提出基于参考文本的LLM创造力自动评估方法，显著提升与人类评估的一致性。 large language model
6 Cequel: Cost-Effective Querying of Large Language Models for Text Clustering Cequel：一种低成本的大语言模型文本聚类查询框架 large language model
7 Certified Mitigation of Worst-Case LLM Copyright Infringement 提出BloomScrub以解决LLM版权侵权问题 large language model
8 CAPO: Cost-Aware Prompt Optimization 提出CAPO以提升提示优化的成本效益 large language model
9 Dynamic Early Exit in Reasoning Models 提出动态早期退出机制以提升推理模型效率 chain-of-thought
10 What's the Difference? Supporting Users in Identifying the Effects of Prompt and Model Changes Through Token Patterns Spotlight：通过Token模式分析，辅助用户理解Prompt和模型变更对LLM输出的影响 large language model
11 Performance Evaluation of Emotion Classification in Japanese Using RoBERTa and DeBERTa 利用DeBERTa-v3-large模型实现高精度日语情感分类，并开源模型。 large language model
12 CiteFix: Enhancing RAG Accuracy Through Post-Processing Citation Correction CiteFix：通过后处理引用校正增强RAG系统的准确性 large language model
13 Instruction-Tuning Data Synthesis from Scratch via Web Reconstruction 提出WebR框架，通过Web重建从原始网页中合成高质量指令微调数据 instruction following ✅

🔬 支柱二：RL算法与架构 (RL & Architecture) (7 篇)

#题目一句话要点标签🔗⭐
14 Pre-DPO: Improving Data Utilization in Direct Preference Optimization Using a Guiding Reference Model Pre-DPO：利用引导参考模型提升直接偏好优化中的数据利用率 reinforcement learning RLHF DPO
15 Bidirectional Mamba for Single-Cell Data: Efficient Context Learning with Biological Fidelity GeneMamba：基于双向Mamba的单细胞数据高效上下文学习模型 Mamba state space model foundation model
16 SARI: Structured Audio Reasoning via Curriculum-Guided Reinforcement Learning SARI：通过课程引导强化学习实现结构化音频推理 reinforcement learning curriculum learning large language model
17 TTRL: Test-Time Reinforcement Learning 提出TTRL：一种无需标签的测试时强化学习方法，用于提升大语言模型推理能力。 reinforcement learning large language model ✅
18 Honey, I Shrunk the Language Model: Impact of Knowledge Distillation Methods on Performance and Explainability 提出基于批判-修正提示的知识蒸馏方法，提升小模型性能与可解释性 distillation large language model
19 LongMamba: Enhancing Mamba's Long Context Capabilities via Training-Free Receptive Field Enlargement LongMamba：通过免训练感受野扩展增强Mamba的长上下文能力 Mamba SSM state space model ✅
20 Capturing Symmetry and Antisymmetry in Language Models through Symmetry-Aware Training Objectives 提出对称感知训练目标，提升语言模型对对称与反对称关系的理解能力 contrastive learning large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页