Image de couverture de livre

Post Reasoning: Improving the Performance of Non-Thinking Models at No Cost

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Xuan, Richmond Sin Jing, Bhardwaj, Rishabh, Poria, Soujanya
Format:	Preprint
Publié:	2026
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

Language Models are Homer Simpson! Safety Re-Alignment of Fine-tuned Language Models through Task Arithmetic
par: Bhardwaj, Rishabh, et autres
Publié: (2024)

OffTopicEval: When Large Language Models Enter the Wrong Chat, Almost Always!
par: Lei, Jingdi, et autres
Publié: (2025)

Towards Robust Instruction Tuning on Multimodal Large Language Models
par: Han, Wei, et autres
Publié: (2024)

The Jumping Reasoning Curve? Tracking the Evolution of Reasoning Performance in GPT-[n] and o-[n] Models on Multimodal Puzzles
par: Toh, Vernon Y. H., et autres
Publié: (2025)

Uncovering Cross-Linguistic Disparities in LLMs using Sparse Autoencoders
par: Xuan, Richmond Sin Jing, et autres
Publié: (2025)

DELLA-Merging: Reducing Interference in Model Merging through Magnitude-Based Sampling
par: Deep, Pala Tej, et autres
Publié: (2024)

Ruby Teaming: Improving Quality Diversity Search with Memory for Automated Red Teaming
par: Han, Vernon Toh Yan, et autres
Publié: (2024)

Error Typing for Smarter Rewards: Improving Process Reward Models with Error-Aware Hierarchical Supervision
par: Pala, Tej Deep, et autres
Publié: (2025)

Video2Music: Suitable Music Generation from Videos using an Affective Multimodal Transformer model
par: Kang, Jaeyong, et autres
Publié: (2023)

Are Language Models Puzzle Prodigies? Algorithmic Puzzles Unveil Serious Challenges in Multimodal Reasoning
par: Ghosal, Deepanway, et autres
Publié: (2024)

Stacked from One: Multi-Scale Self-Injection for Context Window Extension
par: Han, Wei, et autres
Publié: (2026)

Training Vision-Language Process Reward Models for Test-Time Scaling in Multimodal Reasoning: Key Insights and Lessons Learned
par: Ong, Brandon, et autres
Publié: (2025)

WalledEval: A Comprehensive Safety Evaluation Toolkit for Large Language Models
par: Gupta, Prannaya, et autres
Publié: (2024)

Two are better than one: Context window extension with multi-grained self-injection
par: Han, Wei, et autres
Publié: (2024)

Emma-X: An Embodied Multimodal Action Model with Grounded Chain of Thought and Look-ahead Spatial Reasoning
par: Sun, Qi, et autres
Publié: (2024)

Reasoning Paths Optimization: Learning to Reason and Explore From Diverse Paths
par: Chia, Yew Ken, et autres
Publié: (2024)

Self-Adaptive Sampling for Efficient Video Question-Answering on Image--Text Models
par: Han, Wei, et autres
Publié: (2023)

Can-Do! A Dataset and Neuro-Symbolic Grounded Framework for Embodied Planning with Large Multimodal Models
par: Chia, Yew Ken, et autres
Publié: (2024)

Large Language Models for Automated Open-domain Scientific Hypotheses Discovery
par: Yang, Zonglin, et autres
Publié: (2023)

Toward Robust Multimodal Learning using Multimodal Foundational Models
par: Zhao, Xianbing, et autres
Publié: (2024)

Thinking About Thinking: Evaluating Reasoning in Post-Trained Language Models
par: Singla, Pratham, et autres
Publié: (2025)

Beyond What to Select: A Plug-and-play Oscillatory Data-Volume Scheduling for Efficient Model Training
par: Yang, Suorong, et autres
Publié: (2026)

Thinking-Based Non-Thinking: Solving the Reward Hacking Problem in Training Hybrid Reasoning Models via Reinforcement Learning
par: Gan, Siyuan, et autres
Publié: (2026)

Thinking Sparks!: Emergent Attention Heads in Reasoning Models During Post Training
par: Park, Yein, et autres
Publié: (2025)

NORA: A Small Open-Sourced Generalist Vision Language Action Model for Embodied Tasks
par: Hung, Chia-Yu, et autres
Publié: (2025)

HyperTTS: Parameter Efficient Adaptation in Text to Speech using Hypernetworks
par: Li, Yingting, et autres
Publié: (2024)

Computational Irreducibility as the Foundation of Agency: A Formal Model Connecting Undecidability to Autonomous Behavior in Complex Systems
par: Azadi, Poria
Publié: (2025)

DialogXpert: Driving Intelligent and Emotion-Aware Conversations through Online Value-Based Reinforcement Learning with LLM Priors
par: Rakib, Tazeek Bin Abdur, et autres
Publié: (2025)

Lightweight Spatial Modeling for Combinatorial Information Extraction From Documents
par: Dong, Yanfei, et autres
Publié: (2024)

$δ$-mem: Efficient Online Memory for Large Language Models
par: Lei, Jingdi, et autres
Publié: (2026)

NORA-1.5: A Vision-Language-Action Model Trained using World Model- and Action-based Preference Rewards
par: Hung, Chia-Yu, et autres
Publié: (2025)

Don't Think Longer, Think Wisely: Optimizing Thinking Dynamics for Large Reasoning Models
par: An, Sohyun, et autres
Publié: (2025)

Ferret: Faster and Effective Automated Red Teaming with Reward-Based Scoring Technique
par: Pala, Tej Deep, et autres
Publié: (2024)

JAM: A Tiny Flow-based Song Generator with Fine-grained Controllability and Aesthetic Alignment
par: Liu, Renhang, et autres
Publié: (2025)

10 Open Challenges Steering the Future of Vision-Language-Action Models
par: Poria, Soujanya, et autres
Publié: (2025)

MOOSE-Chem: Large Language Models for Rediscovering Unseen Chemistry Scientific Hypotheses
par: Yang, Zonglin, et autres
Publié: (2024)

PivotRL: High Accuracy Agentic Post-Training at Low Compute Cost
par: Yi, Junkeun, et autres
Publié: (2026)

LLMs Can't Handle Peer Pressure: Crumbling under Multi-Agent Social Interactions
par: Song, Maojia, et autres
Publié: (2025)

Efficient Reasoning with Hidden Thinking
par: Shen, Xuan, et autres
Publié: (2025)

Think First, Diffuse Fast: Improving Diffusion Language Model Reasoning via Autoregressive Plan Conditioning
par: Sauver, Earl J St
Publié: (2026)