Imagen de Portada

Beyond Semantics: The Unreasonable Effectiveness of Reasonless Intermediate Tokens

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Valmeekam, Karthik, Palod, Vardhan, Stechly, Kaya, Gundawar, Atharva, Kambhampati, Subbarao
Formato:	Preprint
Publicado:	2025
Materias:	Machine Learning Artificial Intelligence
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Performative Thinking? The Brittle Correlation Between CoT Length and Problem Complexity
por: Palod, Vardhan, et al.
Publicado: (2025)

Planning in Strawberry Fields: Evaluating and Improving the Planning and Scheduling Capabilities of LRM o1
por: Valmeekam, Karthik, et al.
Publicado: (2024)

Chain of Thoughtlessness? An Analysis of CoT in Planning
por: Stechly, Kaya, et al.
Publicado: (2024)

On the Self-Verification Limitations of Large Language Models on Reasoning and Planning Tasks
por: Stechly, Kaya, et al.
Publicado: (2024)

LLMs Still Can't Plan; Can LRMs? A Preliminary Evaluation of OpenAI's o1 on PlanBench
por: Valmeekam, Karthik, et al.
Publicado: (2024)

Position: Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces!
por: Kambhampati, Subbarao, et al.
Publicado: (2025)

RL in Name Only? Analyzing the Structural Assumptions in RL post-training for LLMs
por: Samineni, Soumya Rani, et al.
Publicado: (2025)

Robust Planning with Compound LLM Architectures: An LLM-Modulo Approach
por: Gundawar, Atharva, et al.
Publicado: (2024)

LLMs Can't Plan, But Can Help Planning in LLM-Modulo Frameworks
por: Kambhampati, Subbarao, et al.
Publicado: (2024)

Robust Planning with LLM-Modulo Framework: Case Study in Travel Planning
por: Gundawar, Atharva, et al.
Publicado: (2024)

(How) Do reasoning models reason?
por: Subbarao Kambhampati, et al.
Publicado: (2025)

Who is Helping Whom? Analyzing Inter-dependencies to Evaluate Cooperation in Human-AI Teaming
por: Biswas, Upasana, et al.
Publicado: (2025)

Can Large Language Models Reason and Plan?
por: Kambhampati, Subbarao
Publicado: (2024)

Mind The Gap: Quantifying Mechanistic Gaps in Algorithmic Reasoning via Neural Compilation
por: Saldyt, Lucas, et al.
Publicado: (2025)

Superior Computer Chess with Model Predictive Control, Reinforcement Learning, and Rollout
por: Gundawar, Atharva, et al.
Publicado: (2024)

Evaluating the False Trust Engendered by LLM Explanations
por: Palod, Vardhan, et al.
Publicado: (2026)

The Unreasonable Effectiveness of Entropy Minimization in LLM Reasoning
por: Agarwal, Shivam, et al.
Publicado: (2025)

The Unreasonable Effectiveness of Eccentric Automatic Prompts
por: Battle, Rick, et al.
Publicado: (2024)

Local Coherence or Global Validity? Investigating RLVR Traces in Math Domains
por: Samineni, Soumya Rani, et al.
Publicado: (2025)

Extracting Heuristics from Large Language Models for Reward Shaping in Reinforcement Learning
por: Bhambri, Siddhant, et al.
Publicado: (2024)

Incorporating Human Flexibility through Reward Preferences in Human-AI Teaming
por: Bhambri, Siddhant, et al.
Publicado: (2023)

The Unreasonable Effectiveness of Easy Training Data for Hard Tasks
por: Hase, Peter, et al.
Publicado: (2024)

The Unreasonable Effectiveness of Open Science in AI: A Replication Study
por: Gundersen, Odd Erik, et al.
Publicado: (2024)

The Unreasonable Effectiveness of Discrete-Time Gaussian Process Mixtures for Robot Policy Learning
por: von Hartz, Jan Ole, et al.
Publicado: (2025)

Beyond Semantic Manipulation: Token-Space Attacks on Reward Models
por: Zhang, Yuheng, et al.
Publicado: (2026)

The Unreasonable Effectiveness of Model Merging for Cross-Lingual Transfer in LLMs
por: Bandarkar, Lucas, et al.
Publicado: (2025)

PAC Bench: Do Foundation Models Understand Prerequisites for Executing Manipulation Policies?
por: Gundawar, Atharva, et al.
Publicado: (2025)

Algorithmic Language Models with Neurally Compiled Libraries
por: Saldyt, Lucas, et al.
Publicado: (2024)

Explainable AI For Early Detection Of Sepsis
por: Thakur, Atharva, et al.
Publicado: (2025)

Explainable Human-AI Interaction: A Planning Perspective
por: Sreedharan, Sarath, et al.
Publicado: (2024)

GenPlan: Generative Sequence Models as Adaptive Planners
por: Karthikeyan, Akash, et al.
Publicado: (2024)

SeqMate: A Novel Large Language Model Pipeline for Automating RNA Sequencing
por: Mondal, Devam, et al.
Publicado: (2024)

Do Cognitively Interpretable Reasoning Traces Improve LLM Performance?
por: Bhambri, Siddhant, et al.
Publicado: (2025)

Interpretable Traces, Unexpected Outcomes: Investigating the Disconnect in Trace-Based Knowledge Distillation
por: Bhambri, Siddhant, et al.
Publicado: (2025)

Learning from Ambiguous Demonstrations with Self-Explanation Guided Reinforcement Learning
por: Zha, Yantian, et al.
Publicado: (2021)

On the Brittle Foundations of ReAct Prompting for Agentic Large Language Models
por: Verma, Mudit, et al.
Publicado: (2024)

Adaptformer: Sequence models as adaptive iterative planners
por: Karthikeyan, Akash, et al.
Publicado: (2024)

Beyond Multi-Token Prediction: Pretraining LLMs with Future Summaries
por: Mahajan, Divyat, et al.
Publicado: (2025)

Beyond the 80/20 Rule: High-Entropy Minority Tokens Drive Effective Reinforcement Learning for LLM Reasoning
por: Wang, Shenzhi, et al.
Publicado: (2025)

Mid-Think: Training-Free Intermediate-Budget Reasoning via Token-Level Triggers
por: Yang, Wang, et al.
Publicado: (2026)