Consolidating Reasoning with Test-Time Training
cs.CL, cs.AI
Submitted: 2026-11-01
Updated: 2026-11-01
Code: https://github.com/project-numina/aimo-progress-prize
Terminology
Sources
- Titans: Learning to Memorize at Test Time
- ATLAS: Learning to Optimally Memorize the Context at Test Time
- Large Language Monkeys: Scaling Inference Compute with Repeated Sampling
- LSPO: Length-aware Dynamic Sampling for Policy Optimization in LLM Reasoning
- Universal Self-Consistency for Large Language Model Generation
- AceReason-Nemotron: Advancing Math and Code Reasoning through Reinforcement Learning
- PERK: Long-Context Reasoning as Test-Time Learning
- Training Verifiers to Solve Math Word Problems
- DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
- CL-bench: A Benchmark for Context Learning
- Improving Factuality and Reasoning in Language Models through Multiagent Debate
- In-Place Test-Time Training
- Deep Think with Confidence
- Omni-MATH: A Universal Olympiad Level Mathematic Benchmark For Large Language Models
- OPE: Overcoming Information Saturation in Parallel Thinking via Outline-Guided Path Exploration
- RULER: What's the Real Context Size of Your Long-Context Language Models?
- Beyond Log Likelihood: Probability-Based Objectives for Supervised Fine-Tuning across the Model Capability Continuum
- Thoughtbubbles: an Unsupervised Method for Parallel Thinking in Latent Space
- Decoupled Weight Decay Regularization
- Learning Adaptive Parallel Reasoning with Language Models
Related papers
- ALARM: Audio-Language Alignment for Reasoning Models
- ThinkSum: Probabilistic reasoning over sets using large language models
- Coherence boosting: When your pretrained language model is not paying enough attention
- From Facts to Insights: A Persona-Driven Dual Memory Framework and Dataset for Role-Playing Agents
- Improving General Role-Playing Agents via Psychology-Grounded Reasoning and Role-Aware Policy Optimization
- The Art of Asking: Multilingual Prompt Optimization for Synthetic Data