Image de couverture de livre

Bridging Latent Reasoning and Target-Language Generation via Retrieval-Transition Heads

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Patel, Shaswat, Trivedi, Vishvesh, Han, Yue, Hong, Yihuai, Choi, Eunsol
Format:	Preprint
Publié:	2026
Sujets:	Computation and Language
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Beyond Single Embeddings: Capturing Diverse Targets with Multi-Query Retrieval
par: Chen, Hung-Ting, et autres
Publié: (2025)

Open-World Evaluation for Retrieving Diverse Perspectives
par: Chen, Hung-Ting, et autres
Publié: (2024)

CLaRa: Bridging Retrieval and Generation with Continuous Latent Reasoning
par: He, Jie, et autres
Publié: (2025)

RVR: Retrieve-Verify-Retrieve for Comprehensive Question Answering
par: Qian, Deniz, et autres
Publié: (2026)

AmbigDocs: Reasoning across Documents on Different Entities under the Same Name
par: Lee, Yoonsang, et autres
Publié: (2024)

RARe: Retrieval Augmented Retrieval with In-Context Examples
par: Tejaswi, Atula, et autres
Publié: (2024)

Understanding Retrieval Augmentation for Long-Form Question Answering
par: Chen, Hung-Ting, et autres
Publié: (2023)

RefreshKV: Updating Small KV Cache During Long-form Generation
par: Xu, Fangyuan, et autres
Publié: (2024)

Learning to Reason Across Parallel Samples for LLM Reasoning
par: Qi, Jianing, et autres
Publié: (2025)

DiffAdapt: Difficulty-Adaptive Reasoning for Token-Efficient LLM Inference
par: Liu, Xiang, et autres
Publié: (2025)

Complex Claim Verification with Evidence Retrieved in the Wild
par: Chen, Jifan, et autres
Publié: (2023)

The Reasoning-Memorization Interplay in Language Models Is Mediated by a Single Direction
par: Hong, Yihuai, et autres
Publié: (2025)

From Distributional to Overton Pluralism: Investigating Large Language Model Alignment
par: Lake, Thom, et autres
Publié: (2024)

On Language Models' Sensitivity to Suspicious Coincidences
par: Padmanabhan, Sriram, et autres
Publié: (2025)

StockBot 2.0: Vanilla LSTMs Outperform Transformer-based Forecasting for Stock Prices
par: Mohanty, Shaswat
Publié: (2026)

Contrastive Learning to Improve Retrieval for Real-world Fact Checking
par: Sriram, Aniruddh, et autres
Publié: (2024)

Exploring Design Choices for Building Language-Specific LLMs
par: Tejaswi, Atula, et autres
Publié: (2024)

Mitigating Temporal Misalignment by Discarding Outdated Facts
par: Zhang, Michael J. Q., et autres
Publié: (2023)

Hybrid Latent Reasoning via Reinforcement Learning
par: Yue, Zhenrui, et autres
Publié: (2025)

Bridging Relevance and Reasoning: Rationale Distillation in Retrieval-Augmented Generation
par: Jia, Pengyue, et autres
Publié: (2024)

ParaSpeechCLAP: A Dual-Encoder Speech-Text Model for Rich Stylistic Language-Audio Pretraining
par: Diwan, Anuj, et autres
Publié: (2026)

BAT: Learning to Reason about Spatial Sounds with Large Language Models
par: Zheng, Zhisheng, et autres
Publié: (2024)

Not All Heads Matter: A Head-Level KV Cache Compression Method with Integrated Retrieval and Reasoning
par: Fu, Yu, et autres
Publié: (2024)

Retrieval Heads are Dynamic
par: Lin, Yuping, et autres
Publié: (2026)

BRIEF: Bridging Retrieval and Inference for Multi-hop Reasoning via Compression
par: Li, Yuankai, et autres
Publié: (2024)

Precise In-Parameter Concept Erasure in Large Language Models
par: Gur-Arieh, Yoav, et autres
Publié: (2025)

Latent Thoughts Tuning: Bridging Context and Reasoning with Fused Information in Latent Tokens
par: Liu, Weihao, et autres
Publié: (2026)

Rhapsody: A Dataset for Highlight Detection in Podcasts
par: Park, Younghan, et autres
Publié: (2025)

Improving LLM-as-a-Judge Inference with the Judgment Distribution
par: Wang, Victor, et autres
Publié: (2025)

Crafting In-context Examples according to LMs' Parametric Knowledge
par: Lee, Yoonsang, et autres
Publié: (2023)

User Feedback in Human-LLM Dialogues: A Lens to Understand Users But Noisy as a Learning Signal
par: Liu, Yuhan, et autres
Publié: (2025)

Failing to Falsify: Evaluating and Mitigating Confirmation Bias in Language Models
par: Jhaveri, Ayush Rajesh, et autres
Publié: (2026)

Scaling Latent Reasoning via Looped Language Models
par: Zhu, Rui-Jie, et autres
Publié: (2025)

LatentRAG: Latent Reasoning and Retrieval for Efficient Agentic RAG
par: Zheng, Yijia, et autres
Publié: (2026)

The Rise of Parameter Specialization for Knowledge Storage in Large Language Models
par: Hong, Yihuai, et autres
Publié: (2025)

Modeling Future Conversation Turns to Teach LLMs to Ask Clarifying Questions
par: Zhang, Michael J. Q., et autres
Publié: (2024)

Retrieval-Reasoning Large Language Model-based Synthetic Clinical Trial Generation
par: Xu, Zerui, et autres
Publié: (2024)

PCToolkit: A Unified Plug-and-Play Prompt Compression Toolkit of Large Language Models
par: Li, Jinyi, et autres
Publié: (2024)

Tool Retrieval Bridge: Aligning Vague Instructions with Retriever Preferences via Bridge Model
par: Chen, Kunfeng, et autres
Publié: (2026)

Improving Contextual Faithfulness of Large Language Models via Retrieval Heads-Induced Optimization
par: Huang, Lei, et autres
Publié: (2025)