token representation

Follow the Flow: On Information Flow Across Textual Tokens in Text-to-Image Models ACL 2026 Identifying and Analyzing Performance-Critical Tokens in Large Language Models AAAI 2026 Improved Masked Image Generation with Knowledge-Augmented Token Representations AAAI 2026 Linguistic Profiling of Transformer Embedding Geometry CoNLL 2026 Morpheme Matters: Morpheme-Based Subword Tokenization for Korean Language Models EACL 2026 Token-Context Attention for NLI: An Alternative to Self-Attention AAAI 2026 Attention-Only Transformers via Unrolled Subspace Denoising ICML 2025 Breaking Token Into Concepts: Exploring Extreme Compression in Token Representation Via Compositional Shared Semantics EMNLP 2025 Collapsed Language Models Promote Fairness ICLR 2025 Cross-modal Information Flow in Multimodal Large Language Models CVPR 2025 FlowTok: Flowing Seamlessly Across Text and Image Tokens ICCV 2025 GETMusic: Generating Music Tracks with a Unified Representation and Diffusion Framework IJCAI 2025 Inceptive Transformers: Enhancing Contextual Representations through Multi-Scale Feature Learning Across Domains and Languages EMNLP 2025 Padding Tone: A Mechanistic Analysis of Padding Tokens in T2I Models NAACL 2025 Scaling Capability in Token Space: An Analysis of Large Vision Language Model JMLR 2025 ToolGen: Unified Tool Retrieval and Calling via Generation ICLR 2025 Unified Molecule-Text Language Model with Discrete Token Representation IJCAI 2025 A Meta-Learning Perspective on Transformers for Causal Language Modeling ACL 2024 BTR: Binary Token Representations for Efficient Retrieval Augmented Language Models ICLR 2024 Image Understanding Makes for A Good Tokenizer for Image Generation NeurIPS 2024 Mitigating Frequency Bias and Anisotropy in Language Model Pre-Training with Syntactic Smoothing EMNLP 2024 Scaling White-Box Transformers for Vision NeurIPS 2024 SparseFlow: Accelerating Transformers by Sparsifying Information Flows ACL 2024 Successor Heads: Recurring, Interpretable Attention Heads In The Wild ICLR 2024 Synergetic Interaction Network with Cross-task Attention for Joint Relational Triple Extraction COLING 2024