No article text was recorded for this version.
pre-training knowledge ceiling
Version 1 of 166 papers80 conceptsfrozen 2026-10-04
Papers in this version
- Blog: Using Reinforcement Learning to Train Large Language Models to Explain Human Decisions
- Blog: Evaluating the Progression of Large Language Model Capabilities for Small-Molecule Drug Design
- Blog: One SPACE to Rule Them All: Jointly Mitigating Factuality and Faithfulness Hallucinations in LLMs
- Blog: Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model?
- Blog: Expanding Reasoning Potential in Foundation Model by Learning Diverse Chains of Thought Patterns
- Blog: Right Question is Already Half the Answer: Fully Unsupervised LLM Reasoning Incentivization
- Blog: R1-Code-Interpreter: LLMs Reason with Code via Supervised and Multi-stage Reinforcement Learning
- Blog: Abstraction-of-Thought Makes Language Models Better Reasoners
- Blog: Towards Understanding Self-play for LLM Reasoning
- Blog: Can Large Language Models Develop Strategic Reasoning? Post-training Insights from Learning Chess
- Blog: ProRL: Prolonged Reinforcement Learning Expands Reasoning Boundaries in Large Language Models
- Blog: Surrogate Signals from Format and Length: Reinforcement Learning for Solving Mathematical Problems without Ground Truth Answers
- Blog: Using Reinforcement Learning to Train Large Language Models to Explain Human Decisions
- Blog: Masked Distillation: Internalizing the Chain-of-Thought in Language Models
- Blog: Beyond 'Aha!': Toward Systematic Meta-Abilities Alignment in Large Reasoning Models
- Blog: RFG: Test-Time Scaling for Diffusion Large Language Model Reasoning with Reward-Free Guidance
- Blog: Spurious Rewards: Rethinking Training Signals in RLVR
- Blog: Can GRPO Help LLMs Transcend Their Pretraining Origin?
- Blog: Test-Time Scaling with Reflective Generative Model
- Blog: Limits of Generalization in RLVR: Two Case Studies in Mathematical Reasoning
- Blog: Active Layer-Contrastive Decoding Reduces Hallucination in Large Language Model Generation
- Blog: Correcting in Hindsight: Editing Past Key-Value States for Robust LLM Reasoning
- Blog: Dependency Matters: Enhancing LLM Reasoning with Explicit Knowledge Grounding
- Blog: Principled RL for Diffusion LLMs Emerges from a Sequence-Level Perspective
- Blog: DRIFT: Decoupled Rollouts and Importance-Weighted Fine-Tuning for Efficient Multi-Turn Optimization
- Blog: Improve Decoding Factuality by Token-wise Cross Layer Entropy of Large Language Models
- Blog: ALW: Adaptive Layer-Wise contrastive decoding enhancing reasoning ability in Large Language Models
- Blog: Reinforcement Learning for Out-of-Distribution Reasoning in LLMs: An Empirical Study on Diagnosis-Related Group Coding
- Blog: Video Reasoning without Training
- Blog: Discovering Latent Knowledge in Language Models Without Supervision
- Blog: Long-form Hallucination Detection with Self-elicitation
- Blog: Steering LLMs Towards Unbiased Responses: A Causality-Guided Debiasing Framework
- Blog: Principled RL for Diffusion LLMs Emerges from a Sequence-Level Perspective
- Blog: Skills-in-Context Prompting: Unlocking Compositionality in Large Language Models
- Blog: Entropy Guided Extrapolative Decoding to Improve Factuality in Large Language Models
- Blog: Text Before Vision: Staged Knowledge Injection Matters for Agentic RLVR in Ultra-High-Resolution Remote Sensing Understanding
- Blog: Mitigating Forgetting Between Supervised and Reinforcement Learning Yields Stronger Reasoners
- Blog: AutoAct: Automatic Agent Learning from Scratch for QA via Self-Planning
- Blog: Recall-Extend Dynamics: Enhancing Small Language Models through Controlled Exploration and Refined Offline Integration
- Blog: A multi-agent system based on HNC for domain-specific machine translation
- Blog: MetaMath: Bootstrap Your Own Mathematical Questions for Large Language Models
- Blog: metaTextGrad: Automatically optimizing language model optimizers
- Blog: A Collaborative Reasoning Framework Powered by Reinforcement Learning and Large Language Models for Complex Questions Answering over Knowledge Graph
- Blog: InternBootcamp Technical Report: Boosting LLM Reasoning with Verifiable Task Scaling
- Blog: CoFFT: Chain of Foresight-Focus Thought for Visual Language Models
- Blog: Mirror: Multiple-perspective Self-Reflection Method for Knowledge-rich Reasoning
- Blog: Post-Training with Policy Gradients: Optimality and the Base Model Barrier
- Blog: DoLa: Decoding by Contrasting Layers Improves Factuality in Large Language Models
- Blog: Subjective Topic meets LLMs: Unleashing Comprehensive, Reflective and Creative Thinking through the Negation of Negation
- Blog: Understanding Contextual Recall in Transformers: How Finetuning Enables In-Context Reasoning over Pretraining Knowledge
- Blog: ThinkTuning: Instilling Cognitive Reflections without Distillation
- Blog: Evaluating the Progression of Large Language Model Capabilities for Small-Molecule Drug Design
- Blog: You Are My Type! Type Embeddings for Pre-trained Language Models
- Blog: Unified Multimodal Chain-of-Thought Reward Model through Reinforcement Fine-Tuning
- Blog: REAL: Regression-Aware Reinforcement Learning for LLM-as-a-Judge
- Blog: The Debate on RLVR Reasoning Capability Boundary: Shrinkage, Expansion, or Both? A Two-Stage Dynamic View
- Blog: SeRL: Self-play Reinforcement Learning for Large Language Models with Limited Data
- Blog: Whatever Remains Must Be True: Filtering Drives Reasoning in LLMs, Shaping Diversity
- Blog: Self-Improving LLM Agents at Test-Time
- Blog: Irreducible Supervision Enables Compositional Generalization in Post-Training
Showing 60 of 66 papers.
History & citation
Current · Permanent link
Cite this direction
Latest version (a living page):
@misc{lacuna_pre_training_knowledge_ceiling_59d0dcd8,
title = {pre-training knowledge ceiling},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/pre-training-knowledge-ceiling/rdr_59d0dcd8b93949fea8228de1b8132d68},
note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-19},
}This exact version (version 1, frozen text):
@misc{lacuna_pre_training_knowledge_ceiling_59d0dcd8_v1,
title = {pre-training knowledge ceiling},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/pre-training-knowledge-ceiling/rdr_59d0dcd8b93949fea8228de1b8132d68/v/1},
note = {Version 1, frozen 2026-10-04. Accessed 2026-11-19},
}