Image de couverture de livre

Small Models Struggle to Learn from Strong Reasoners

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Li, Yuetai, Yue, Xiang, Xu, Zhangchen, Jiang, Fengqing, Niu, Luyao, Lin, Bill Yuchen, Ramasubramanian, Bhaskar, Poovendran, Radha
Format:	Preprint
Publié:	2025
Sujets:	Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Temporal Sampling for Forgotten Reasoning in LLMs
par: Li, Yuetai, et autres
Publié: (2025)

TinyV: Reducing False Negatives in Verification Improves RL for LLM Reasoning
par: Xu, Zhangchen, et autres
Publié: (2025)

CleanGen: Mitigating Backdoor Attacks for Generation Tasks in Large Language Models
par: Li, Yuetai, et autres
Publié: (2024)

VisualSphinx: Large-Scale Synthetic Vision Logic Puzzles for RL
par: Feng, Yichen, et autres
Publié: (2025)

SafeChain: Safety of Language Models with Long Chain-of-Thought Reasoning Capabilities
par: Jiang, Fengqing, et autres
Publié: (2025)

Stronger Models are NOT Stronger Teachers for Instruction Tuning
par: Xu, Zhangchen, et autres
Publié: (2024)

ChatBug: A Common Vulnerability of Aligned LLMs Induced by Chat Templates
par: Jiang, Fengqing, et autres
Publié: (2024)

ArtPrompt: ASCII Art-based Jailbreak Attacks against Aligned LLMs
par: Jiang, Fengqing, et autres
Publié: (2024)

SafeDecoding: Defending against Jailbreak Attacks via Safety-Aware Decoding
par: Xu, Zhangchen, et autres
Publié: (2024)

SoSBench: Benchmarking Safety Alignment on Six Scientific Domains
par: Jiang, Fengqing, et autres
Publié: (2025)

Magpie: Alignment Data Synthesis from Scratch by Prompting Aligned LLMs with Nothing
par: Xu, Zhangchen, et autres
Publié: (2024)

ACE: A Model Poisoning Attack on Contribution Evaluation Methods in Federated Learning
par: Xu, Zhangchen, et autres
Publié: (2024)

The WidthWall: A Strict Expressivity Hierarchy for Hypergraph Neural Networks
par: Jiang, Fengqing, et autres
Publié: (2026)

BadScientist: Can a Research Agent Write Convincing but Unsound Papers that Fool LLM Reviewers?
par: Jiang, Fengqing, et autres
Publié: (2025)

Brave: Byzantine-Resilient and Privacy-Preserving Peer-to-Peer Federated Learning
par: Xu, Zhangchen, et autres
Publié: (2024)

Polyhedral Instability Governs Regret in Online Learning
par: Li, Yuetai, et autres
Publié: (2026)

Visual Aesthetic Benchmark: Can Frontier Models Judge Beauty?
par: Feng, Yichen, et autres
Publié: (2026)

JobBench: Aligning Agent Work With Human Will
par: Li, Yuetai, et autres
Publié: (2026)

Does Math Reasoning Improve General LLM Capabilities? Understanding Transferability of LLM Reasoning
par: Huan, Maggie, et autres
Publié: (2025)

A Method for Fast Autonomy Transfer in Reinforcement Learning
par: Sahabandu, Dinuka, et autres
Publié: (2024)

Double-Dip: Thwarting Label-Only Membership Inference Attacks with Transfer Learning and Randomization
par: Rajabi, Arezoo, et autres
Publié: (2024)

Fault Tolerant Neural Control Barrier Functions for Robotic Systems under Sensor Faults and Attacks
par: Zhang, Hongchao, et autres
Publié: (2024)

BadChain: Backdoor Chain-of-Thought Prompting for Large Language Models
par: Xiang, Zhen, et autres
Publié: (2024)

Modeling and Designing Non-Pharmaceutical Interventions in Epidemics: A Submodular Approach
par: Cheng, Shiyu, et autres
Publié: (2024)

Simulating Environments with Reasoning Models for Agent Training
par: Li, Yuetai, et autres
Publié: (2025)

Distributed Safety-Critical Control of Multi-Agent Systems with Time-Varying Communication Topologies
par: Cheng, Shiyu, et autres
Publié: (2026)

Swarm-STL: A Framework for Motion Planning in Large-Scale, Multi-Swarm Systems
par: Cheng, Shiyu, et autres
Publié: (2025)

KodCode: A Diverse, Challenging, and Verifiable Synthetic Dataset for Coding
par: Xu, Zhangchen, et autres
Publié: (2025)

Steering Multimodal Large Language Models Decoding for Context-Aware Safety
par: Liu, Zheyuan, et autres
Publié: (2025)

ZebraLogic: On the Scaling Limits of LLMs for Logical Reasoning
par: Lin, Bill Yuchen, et autres
Publié: (2025)

Game of Trojans: Adaptive Adversaries Against Output-based Trojaned-Model Detectors
par: Sahabandu, Dinuka, et autres
Publié: (2024)

Who is Responsible? Explaining Safety Violations in Multi-Agent Cyber-Physical Systems
par: Niu, Luyao, et autres
Publié: (2024)

TOUCAN: Synthesizing 1.5M Tool-Agentic Data from Real-World MCP Environments
par: Xu, Zhangchen, et autres
Publié: (2025)

Long-context LLMs Struggle with Long In-context Learning
par: Li, Tianle, et autres
Publié: (2024)

The Strongest Teacher Is Not Always the Best Teacher: Student-Centric Answer Selection
par: Hu, Zhengyu, et autres
Publié: (2026)

Reasoning Models Struggle to Control their Chains of Thought
par: Yueh-Han, Chen, et autres
Publié: (2026)

CANTXSec: A Deterministic Intrusion Detection and Prevention System for CAN Bus Monitoring ECU Activations
par: Donadel, Denis, et autres
Publié: (2025)

Collaborative Agent Reasoning Engineering (CARE): A Three-Party Design Methodology for Systematically Engineering AI Agents with Subject Matter Experts, Developers, and Helper Agents
par: Ramachandran, Rahul, et autres
Publié: (2026)

Speculative Thinking: Enhancing Small-Model Reasoning with Large Model Guidance at Inference Time
par: Yang, Wang, et autres
Publié: (2025)

Trial and Error: Exploration-Based Trajectory Optimization for LLM Agents
par: Song, Yifan, et autres
Publié: (2024)