LLM Papers

OptiMer: Optimal Distribution Vector Merging Is Better than Data Mixing for Continual Pre-Training Haiyue Song, Masao Utiyama arxiv.org/abs/2603.28858 [𝚌𝚜.𝙲𝙻 𝚌𝚜.𝙰𝙸 𝚌𝚜.𝙻𝙶]

From Consensus to Split Decisions: ABC-Stratified Sentiment in Holocaust Oral Histories Daban Q. Jaff arxiv.org/abs/2603.28913 [𝚌𝚜.𝙲𝙻]

CrossTrace: A Cross-Domain Dataset of Grounded Scientific Reasoning Traces for Hypothesis Generation Andrew Bouras, OMS-II Research Fellow arxiv.org/abs/2603.28924 [𝚌𝚜.𝙲𝙻]

Theory of Mind and Self-Attributions of Mentality are Dissociable in LLMs Junsol Kim, Winnie Street, Roberta Rocca, Daine M. Korngiebel, Adam Waytz, James Evans, Geoff Keeling arxiv.org/abs/2603.28925 [𝚌𝚜.𝙲𝙻 𝚌𝚜.𝙰𝙸]

Known Intents, New Combinations: Clause-Factorized Decoding for Compositional Multi-Intent Detection Abhilash Nandy arxiv.org/abs/2603.28929 [𝚌𝚜.𝙲𝙻]

Human-Like Lifelong Memory: A Neuroscience-Grounded Architecture for Infinite Interaction Diego C. Lerma-Torres (Universidad de Guanajuato) arxiv.org/abs/2603.29023 [𝚌𝚜.𝙲𝙻 𝚌𝚜.𝙰𝙸]

The Model Says Walk: How Surface Heuristics Override Implicit Constraints in LLM Reasoning Yubo Li, Lu Zhang, Tianchong Jiang, Ramayya Krishnan, Rema Padman arxiv.org/abs/2603.29025 [𝚌𝚜.𝙲𝙻 𝚌𝚜.𝙰𝙸]

On the limited utility of parallel data for learning shared multilingual representations Julius Leino, Jörg Tiedemann arxiv.org/abs/2603.29026 [𝚌𝚜.𝙲𝙻]

An Empirical Recipe for Universal Phone Recognition Shikhar Bharadwaj, Chin-Jou Li, Kwanghee Choi, Eunjung Yeo, William Chen, Shinji Watanabe, David R. Mortensen arxiv.org/abs/2603.29042 [𝚌𝚜.𝙲𝙻]

Dual Perspectives in Emotion Attribution: A Generator-Interpreter Framework for Cross-Cultural Analysis of Emotion in LLMs Aizirek Turdubaeva, Uichin Lee arxiv.org/abs/2603.29077 [𝚌𝚜.𝙲𝙻]

PolarQuant: Optimal Gaussian Weight Quantization via Hadamard Rotation for LLM Compression Caio Vicentino arxiv.org/abs/2603.29078 [𝚌𝚜.𝙲𝙻 𝚌𝚜.𝙻𝙶]

APEX-EM: Non-Parametric Online Learning for Autonomous Agents via Structured Procedural-Episodic Experience Replay Pratyay Banerjee, Masud Moshtaghi, Ankit Chadha arxiv.org/abs/2603.29093 [𝚌𝚜.𝙲𝙻 𝚌𝚜.𝙰𝙸 𝚌𝚜.𝙸𝚁]

Concept Training for Human-Aligned Language Models Christine Zhang, Dan Jurafsky, Chen Shani arxiv.org/abs/2603.29123 [𝚌𝚜.𝙲𝙻]

Kwame 2.0: Human-in-the-Loop Generative AI Teaching Assistant for Large Scale Online Coding Education in Africa George Boateng, Samuel Boateng, Victor Kumbol arxiv.org/abs/2603.29159 [𝚌𝚜.𝙲𝙻 𝚌𝚜.𝙲𝚈 𝚌𝚜.𝙷𝙲]

SyriSign: A Parallel Corpus for Arabic Text to Syrian Arabic Sign Language Translation Mohammad Amer Khalil, Raghad Nahas, Ahmad Nassar, Khloud Al Jallad arxiv.org/abs/2603.29219 [𝚌𝚜.𝙲𝙻 𝚌𝚜.𝙰𝙸 𝚌𝚜.𝙲𝚅 𝚌𝚜.𝙷𝙲]

SiPaKosa: A Comprehensive Corpus of Canonical and Classical Buddhist Texts in Sinhala and Pali Ranidu Gurusinghe, Nevidu Jayatilleke arxiv.org/abs/2603.29221 [𝚌𝚜.𝙲𝙻]

Long-Document QA with Chain-of-Structured-Thought and Fine-Tuned SLMs Zhuowen Liang, Xiaotian Lin, Zhengxuan Zhang, Yuyu Luo, Haixun Wang, Nan Tang arxiv.org/abs/2603.29232 [𝚌𝚜.𝙲𝙻 𝚌𝚜.𝙰𝙸 𝚌𝚜.𝙻𝙶]

The Thiomi Dataset: A Large-Scale Multimodal Corpus for Low-Resource African Languages Hillary Mutisya, John Mugane, Gavin Nyamboga, Brian Chege, Maryruth Gathoni arxiv.org/abs/2603.29244 [𝚌𝚜.𝙲𝙻 𝚌𝚜.𝙻𝙶]

MemRerank: Preference Memory for Personalized Product Reranking Zhiyuan Peng, Xuyang Wu, Huaixiao Tou, Yi Fang, Yi Gong arxiv.org/abs/2603.29247 [𝚌𝚜.𝙲𝙻 𝚌𝚜.𝙰𝙸 𝚌𝚜.𝙻𝙶]

CADEL: A Corpus of Administrative Web Documents for Japanese Entity Linking Shohei Higashiyama, Masao Ideuchi, Masao Utiyama arxiv.org/abs/2603.29336 [𝚌𝚜.𝙲𝙻]