Image de couverture de livre

How Do Latent Reasoning Methods Perform Under Weak and Strong Supervision?

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Cui, Yingqian, Dai, Zhenwei, He, Bing, Shi, Zhan, Liu, Hui, Sun, Rui, Liu, Zhiji, Xing, Yue, Tang, Jiliang, Dumoulin, Benoit
Format:	Preprint
Publié:	2026
Sujets:	Artificial Intelligence Computation and Language Machine Learning
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Adaptive Test-Time Reasoning via Reward-Guided Dual-Phase Search
par: Cui, Yingqian, et autres
Publié: (2025)

Towards the Effect of Examples on In-Context Learning: A Theoretical Case Study
par: He, Pengfei, et autres
Publié: (2024)

TRAJECT-Bench:A Trajectory-Aware Benchmark for Evaluating Agentic Tool Use
par: He, Pengfei, et autres
Publié: (2025)

Superiority of Multi-Head Attention in In-Context Linear Regression
par: Cui, Yingqian, et autres
Publié: (2024)

Stepwise Perplexity-Guided Refinement for Efficient Chain-of-Thought Reasoning in Large Language Models
par: Cui, Yingqian, et autres
Publié: (2025)

A Theoretical Understanding of Chain-of-Thought: Coherent Reasoning and Error-Aware Demonstration
par: Cui, Yingqian, et autres
Publié: (2024)

Six-CD: Benchmarking Concept Removals for Benign Text-to-image Diffusion Models
par: Ren, Jie, et autres
Publié: (2024)

DiffusionShield: A Watermark for Copyright Protection against Generative Diffusion Models
par: Cui, Yingqian, et autres
Publié: (2023)

SoK: Machine Unlearning for Large Language Models
par: Ren, Jie, et autres
Publié: (2025)

Weak-to-Strong Reasoning
par: Yang, Yuqing, et autres
Publié: (2024)

Incentivizing Strong Reasoning from Weak Supervision
par: Yuan, Yige, et autres
Publié: (2025)

EnTruth: Enhancing the Traceability of Unauthorized Dataset Usage in Text-to-image Diffusion Models with Minimal and Robust Alterations
par: Ren, Jie, et autres
Publié: (2024)

Examples as the Prompt: A Scalable Approach for Efficient LLM Adaptation in E-Commerce
par: Zeng, Jingying, et autres
Publié: (2025)

FT-Shield: A Watermark Against Unauthorized Fine-tuning in Text-to-Image Diffusion Models
par: Cui, Yingqian, et autres
Publié: (2023)

Position: Agentic Evolution is the Path to Evolving LLMs
par: Lin, Minhua, et autres
Publié: (2026)

Seeing but Not Believing: Probing the Disconnect Between Visual Attention and Answer Correctness in VLMs
par: Liu, Zhining, et autres
Publié: (2025)

Sharpness-Aware Data Poisoning Attack
par: He, Pengfei, et autres
Publié: (2023)

Image-Feature Weak-to-Strong Consistency: An Enhanced Paradigm for Semi-Supervised Learning
par: Wu, Zhiyu, et autres
Publié: (2024)

Weak to Strong: VLM-Based Pseudo-Labeling as a Weakly Supervised Training Strategy in Multimodal Video-based Hidden Emotion Understanding Tasks
par: Wang, Yufei, et autres
Publié: (2026)

CloudMatch: Weak-to-Strong Consistency Learning for Semi-Supervised Cloud Detection
par: Zhao, Jiayi, et autres
Publié: (2026)

A Robust Semantics-based Watermark for Large Language Model against Paraphrasing
par: Ren, Jie, et autres
Publié: (2023)

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

Stealthy Backdoor Attack via Confidence-driven Sampling
par: He, Pengfei, et autres
Publié: (2023)

Co-Supervised Learning: Improving Weak-to-Strong Generalization with Hierarchical Mixture of Experts
par: Liu, Yuejiang, et autres
Publié: (2024)

Evaluating Baseline Correction in Raman Spectroscopy: A Novel Cross‐Wavelength Correlation Approach for Low‐Maturity Organic Matter
par: Zhiji Ou, et autres
Publié: (2025)

Crafting Reversible SFT Behaviors in Large Language Models
par: Lin, Yuping, et autres
Publié: (2026)

Faster Accelerated First-order Methods for Convex Optimization with Strongly Convex Function Constraints
par: Lin, Zhenwei, et autres
Publié: (2022)

Weakly Supervised 3D Open-vocabulary Segmentation
par: Liu, Kunhao, et autres
Publié: (2023)

Weak-to-Strong Preference Optimization: Stealing Reward from Weak Aligned Model
par: Zhu, Wenhong, et autres
Publié: (2024)

How Well Do Self-Supervised Methods Perform in Cross-Domain Few-Shot Learning?
par: Zhang, Yiyi, et autres
Publié: (2022)

Weak-Driven Learning: How Weak Agents make Strong Agents Stronger
par: Chen, Zehao, et autres
Publié: (2026)

The Alignment Floor: How Persona Customization Breaks Safety in Weakly-Aligned LLMs
par: Zhang, Xing, et autres
Publié: (2026)

Weak synchronization in heterogeneous multi-agent systems
par: Stoorvogel, Anton A., et autres
Publié: (2024)

To trust or not to trust: Attention-based Trust Management for LLM Multi-Agent Systems
par: He, Pengfei, et autres
Publié: (2025)

Data Poisoning for In-context Learning
par: He, Pengfei, et autres
Publié: (2024)

Bronchovascular Tree-Guided Weakly Supervised Learning Method for Pulmonary Segment Segmentation
par: Zhao, Ruijie, et autres
Publié: (2025)

Disentangling Latent Shifts of In-Context Learning with Weak Supervision
par: Jukić, Josip, et autres
Publié: (2024)

On Learning Latent Models with Multi-Instance Weak Supervision
par: Wang, Kaifu, et autres
Publié: (2023)

Do Large Language Models Latently Perform Multi-Hop Reasoning?
par: Yang, Sohee, et autres
Publié: (2024)

Weak-to-Strong Generalization beyond Accuracy: a Pilot Study in Safety, Toxicity, and Legal Reasoning
par: Ye, Ruimeng, et autres
Publié: (2024)