Language Modeling

ComPO: Community Preferences for Language Model Personalization NAACL 2025 Confounding Factors in Relating Model Performance to Morphology EMNLP 2025 Constructions are Revealed in Word Distributions EMNLP 2025 Context-Parametric Inversion: Why Instruction Finetuning May Not Actually Improve Context Reliance ICLR 2025 Context Steering: Controllable Personalization at Inference Time ICLR 2025 Convergence and Divergence of Language Models under Different Random Seeds EMNLP 2025 Core Context Aware Transformers for Long Context Language Modeling ICML 2025 Correcting the Mythos of KL-Regularization: Direct Alignment without Overoptimization via Chi-Squared Preference Optimization ICLR 2025 Counterfactual Token Generation in Large Language Models CLeaR 2025 Critic-CoT: Boosting the Reasoning Abilities of Large Language Model via Chain-of-Thought Critic ACL 2025 CSEval: Towards Automated, Multi-Dimensional, and Reference-Free Counterspeech Evaluation using Auto-Calibrated LLMs NAACL 2025 Cultural Alien Sampler: Open-ended Art Generation Balancing Originality and Coherence NeurIPS 2025 Curse of High Dimensionality Issue in Transformer for Long Context Modeling ICML 2025 Customizing the Inductive Biases of Softmax Attention using Structured Matrices ICML 2025 CYCLE-INSTRUCT: Fully Seed-Free Instruction Tuning via Dual Self-Training and Cycle Consistency EMNLP 2025 D2CS - Documents Graph Clustering using LLM supervision EMNLP 2025 DadmaTools V2: an Adapter-Based Natural Language Processing Toolkit for the Persian Language COLING 2025 DASA-Trans-STM: Adaptive Efficient Transformer for Short Text Matching using Data Augmentation and Semantic Awareness EMNLP 2025 Data Drives Unstable Hierarchical Generalization in LMs EMNLP 2025 Data-Efficient Selection via Grammatical Complexity in Continual Pre-training of Domain-Specific LLMs EMNLP 2025 Data with High and Consistent Preference Difference Are Better for Reward Model AAAI 2025 DC-Spin: A Speaker-invariant Speech Tokenizer for Spoken Language Models Interspeech 2025 Debate4MATH: Multi-Agent Debate for Fine-Grained Reasoning in Math ACL 2025 Decoder-Only LLMs can be Masked Auto-Encoders ACL 2025 Decoding Uncertainty: The Impact of Decoding Strategies for Uncertainty Estimation in Large Language Models EMNLP 2025