Imagen de Portada

Pair-In, Pair-Out: Latent Multi-Token Prediction for Efficient LLMs

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Tan, Wenhui, Li, Minghao, Ma, Xiaoqian, Fan, Siqi, Huang, Xiusheng, Zhang, Liujie, Song, Ruihua, Chen, Weihang
Formato:	Preprint
Publicado:	2026
Materias:	Computation and Language Artificial Intelligence
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Hint Tuning: Less Data Makes Better Reasoners
por: Fan, Siqi, et al.
Publicado: (2026)

Commonsense Knowledge Editing Based on Free-Text in LLMs
por: Huang, Xiusheng, et al.
Publicado: (2024)

BFS-PO: Best-First Search for Large Reasoning Models
por: Parascandolo, Fiorenzo, et al.
Publicado: (2026)

If an LLM Were a Character, Would It Know Its Own Story? Evaluating Lifelong Learning in LLMs
por: Fan, Siqi, et al.
Publicado: (2025)

Beyond QA Pairs: Assessing Parameter-Efficient Fine-Tuning for Fact Embedding in LLMs
por: Ratnakar, Shivam, et al.
Publicado: (2025)

MPPO: Multi Pair-wise Preference Optimization for LLMs with Arbitrary Negative Samples
por: Xie, Shuo, et al.
Publicado: (2024)

TokenTiming: A Dynamic Alignment Method for Universal Speculative Decoding Model Pairs
por: Xiao, Sibo, et al.
Publicado: (2025)

Satisfiability Solving with LLMs: A Matched-Pair Evaluation of Reasoning Capability
por: Zhang, Leizhen, et al.
Publicado: (2026)

Reasons and Solutions for the Decline in Model Performance after Editing
por: Huang, Xiusheng, et al.
Publicado: (2024)

Parity-Aware Byte-Pair Encoding: Improving Cross-lingual Fairness in Tokenization
por: Foroutan, Negar, et al.
Publicado: (2025)

Toward Consistent World Models with Multi-Token Prediction and Latent Semantic Enhancement
por: Zhong, Qimin, et al.
Publicado: (2026)

NTPP: Generative Speech Language Modeling for Dual-Channel Spoken Dialogue via Next-Token-Pair Prediction
por: Wang, Qichao, et al.
Publicado: (2025)

Efficient Switchable Safety Control in LLMs via Magic-Token-Guided Co-Training
por: Si, Jianfeng, et al.
Publicado: (2025)

CDTP: A Large-Scale Chinese Data-Text Pair Dataset for Comprehensive Evaluation of Chinese LLMs
por: Wu, Chengwei, et al.
Publicado: (2025)

From Pixels to Tokens: Byte-Pair Encoding on Quantized Visual Modalities
por: Zhang, Wanpeng, et al.
Publicado: (2024)

SelecTKD: Selective Token-Weighted Knowledge Distillation for LLMs
por: Huang, Haiduo, et al.
Publicado: (2025)

TECP: Token-Entropy Conformal Prediction for LLMs
por: Xu, Beining, et al.
Publicado: (2025)

Training LLMs Beyond Next Token Prediction -- Filling the Mutual Information Gap
por: Yang, Chun-Hao, et al.
Publicado: (2025)

ToBlend: Token-Level Blending With an Ensemble of LLMs to Attack AI-Generated Text Detection
por: Huang, Fan, et al.
Publicado: (2024)

AuPair: Golden Example Pairs for Code Repair
por: Mavalankar, Aditi, et al.
Publicado: (2025)

Q-Mirror: Unlocking the Multi-Modal Potential of Scientific Text-Only QA Pairs
por: Wang, Junying, et al.
Publicado: (2025)

OpenSanctions Pairs: Large-Scale Entity Matching with LLMs
por: Smith, Chandler, et al.
Publicado: (2026)

Paired by the Teacher: Turning Unpaired Data into High-Fidelity Pairs for Low-Resource Text Generation
por: Lu, Yen-Ju, et al.
Publicado: (2025)

Semantic Flow Regularization: Teaching LLMs to Generate Diverse Yet Coherent Responses
por: Peng, Kerui, et al.
Publicado: (2026)

DCRM: A Heuristic to Measure Response Pair Quality in Preference Optimization
por: Huang, Chengyu, et al.
Publicado: (2025)

Multi-Stage Verification-Centric Framework for Mitigating Hallucination in Multi-Modal RAG
por: Chen, Baiyu, et al.
Publicado: (2025)

Context Pruning for Coding Agents via Multi-Rubric Latent Reasoning
por: Wang, Jingjing, et al.
Publicado: (2026)

MedSynth: Realistic, Synthetic Medical Dialogue-Note Pairs
por: Mianroodi, Ahmad Rezaie, et al.
Publicado: (2025)

LRHP: Learning Representations for Human Preferences via Preference Pairs
por: Wang, Chenglong, et al.
Publicado: (2024)

Less is More: A Simple yet Effective Token Reduction Method for Efficient Multi-modal LLMs
por: Song, Dingjie, et al.
Publicado: (2024)

Moira: Language-driven Hierarchical Reinforcement Learning for Pair Trading
por: Giannouris, Polydoros, et al.
Publicado: (2026)

Paired Completion: Flexible Quantification of Issue-framing at Scale with LLMs
por: Angus, Simon D, et al.
Publicado: (2024)

Code-Based English Models Surprising Performance on Chinese QA Pair Extraction Task
por: Zheng, Linghan, et al.
Publicado: (2024)

Compressing Sequences in the Latent Embedding Space: $K$-Token Merging for Large Language Models
por: Xu, Zihao, et al.
Publicado: (2026)

Pairing Analogy-Augmented Generation with Procedural Memory for Procedural Q&A
por: Roth, K, et al.
Publicado: (2024)

Efficient Latent Semantic Clustering for Scaling Test-Time Computation of LLMs
por: Lee, Sungjae, et al.
Publicado: (2025)

Automatic Pair Construction for Contrastive Post-training
por: Xu, Canwen, et al.
Publicado: (2023)

Scaling Inference-Efficient Language Models
por: Bian, Song, et al.
Publicado: (2025)

From Implicit to Explicit: Token-Efficient Logical Supervision for Mathematical Reasoning in LLMs
por: Wang, Shaojie, et al.
Publicado: (2026)

PIN: A Knowledge-Intensive Dataset for Paired and Interleaved Multimodal Documents
por: Wang, Junjie, et al.
Publicado: (2024)