arXivDaily arXiv每日学术速递 周一至周五更新

期刊&会议

Annual Meeting of the Association for Computational Linguistics · 会议 · Natural Language Processing

2026-08-11 至 2026-08-11 共收录 7
2608.08829 2026-08-11 cs.CL cs.AI 新提交

Deployable Per-Instance Multi-Layer Activation Steering for Large Language Models

面向大型语言模型的可部署实例级多层激活调控

Muhammad Faishal Adly Nelwan, Alfan Farizki Wicaksono

AI总结 本研究提出可部署的实例级多层激活调控方案,解决现有全局固定层调控的缺陷,在两个8B模型和六个人格特质任务上,实现接近先验的性能,且避免流畅性崩溃。

Comments 43 pages, 24 figures, 30 tables. Under review at ACL Rolling Review (August 2026)

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.17653 2026-08-11 cs.CL

Differences in Typological Alignment in Language Models' Treatment of Differential Argument Marking

语言模型处理差异论元标记中的类型学对齐差异

Iskar Deng, Nathalia Xu, Shane Steinert-Threlkeld

机构 * University of Washington(华盛顿大学)

AI总结 通过控制合成语料训练GPT-2模型,发现模型在标记方向(自然标记方向)上表现出类人偏好,但在论元角色偏好(宾语vs主语)上未复现人类语言的强烈宾语偏好。

Comments 16 pages, 8 figures, 7 tables. To appear at CoNLL 2026

Journal ref Proceedings of the 30th Conference on Computational Natural Language Learning (CoNLL 2026), pp. 268-283, San Diego, California, USA, Association for Computational Linguistics, 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2503.17239 2026-08-11 cs.CL cs.AI

SafeMERGE: Preserving Safety Alignment in Fine-Tuned Large Language Models via Selective Layer-Wise Model Merging

SafeMERGE: 通过选择性层间模型融合在微调大语言模型中保持安全对齐

Aladin Djuhera, Swanand Ravindra Kadhe, Farhan Ahmed, Syed Zawad, Holger Boche

机构 * Technical University Munich(慕尼黑技术大学) IBM Research(IBM研究院)

AI总结 本文提出SafeMERGE,一种轻量级后微调框架,通过选择性融合层间模型来恢复安全对齐,同时保持下游性能,减少有害输出并提升实用性。

Journal ref Findings of the ACL 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2604.11741 2026-08-11 cs.AI 版本更新

Collaborative Multi-Agent Scripts Generation for Enhancing Imperfect-Information Reasoning in Murder Mystery Games

协作多智能体脚本生成以增强谋杀谜游戏中的不完全信息推理

Keyang Zhong, Junlin Xie, Hefeng Wu, Haofeng Li, Guanbin Li

机构 * Sun Yat-sen University(中山大学) The Chinese University of Hong Kong, Shenzhen(香港中文大学(深圳))

AI总结 本文提出协作多智能体框架,通过生成丰富多模态上下文提升VLMs在叙事推理和欺骗鲁棒性中的表现,解决多玩家游戏中不完全信息下的复杂推理问题。

Comments 9 pages, 5 figures, Findings of ACL 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2507.03005 2026-08-11 cs.CL q-bio.PE 版本更新

Beyond cognacy

超越词源关系

Gerhard Jäger

机构 * University of Tübingen(蒂宾根大学)

AI总结 本文比较了传统词源集方法与两种自动方法在语言谱系分析中的效果,发现基于MSA的方法能产生更一致的树结构,预测 typological 变异更准确,具有更清晰的谱系信号。

Comments 10 pages, 1 figure. v3: substantially corrected version - retrained pHMM, corrected T-Coffee implementation, method-independent guide tree, full pipeline re-run on current Lexibank data (974 languages, 15 families); all reported numbers changed, qualitative conclusions unchanged (see title footnote). Data and code: doi:10.57754/FDAT.8n2vd-1bq37

Journal ref Proceedings of the 7th Workshop on Research in Computational Linguistic Typology and Multilingual NLP (SIGTYP 2025), Association for Computational Linguistics

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.03115 2026-08-11 cs.CL eess.AS 版本更新

Discovering and Causally Validating Emotion-Sensitive Neurons in Large Audio-Language Models

在大型音频-语言模型中发现并因果验证情绪敏感神经元

Xiutian Zhao, Björn Schuller, Berrak Sisman

机构 * Center for Language and Speech Processing (CLSP)(语言与语音处理中心) Johns Hopkins University(约翰霍普金斯大学) Group on Language, Audio & Music (GLAM)(语言、音频与音乐小组) Imperial College London(伦敦帝国学院)

AI总结 本研究通过神经元层面的干预验证了大型音频-语言模型中情绪敏感神经元的存在,并揭示了情绪识别的因果机制。

Comments Accepted to ACL 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2506.00549 2026-08-11 cs.CL cs.AI

Towards Multi-dimensional Evaluation of LLM Summarization across Domains and Languages

Hyangsuk Min, Yuho Lee, Minjeong Ban, Jiaqi Deng, Nicole Hee-Yeon Kim, Taewon Yun, Hang Su, Jason Cai, Hwanjun Song

机构 * Korea Advanced Institute of Science and Technology(韩国先进科学研究院) AWS AI Labs(AWS人工智能实验室)

Comments 34 pages, 6 figures

Journal ref ACL 2025 main

详情

展开后加载摘要…

URL PDF HTML 收藏