No article text was recorded for this version.
intermediate difficulty optimization
Version 2 of 271 papers104 conceptsfrozen 2026-10-04
Papers in this version
- Blog: Transferable Curricula through Difficulty Conditioned Generators
- Blog: Staying in the Sweet Spot: Responsive Reasoning Evolution via Capability-Adaptive Hint Scaffolding
- Blog: UltraLogic: Enhancing LLM Reasoning through Large-Scale Data Synthesis and Bipolar Float Reward
- Blog: Online Difficulty Filtering for Reasoning Oriented Reinforcement Learning
- Blog: SPEED-RL: Faster Training of Reasoning Models via Online Curriculum Learning
- Blog: Efficient Reinforcement Finetuning via Adaptive Curriculum Learning
- Blog: Stein Variational Goal Generation for adaptive Exploration in Multi-Goal Reinforcement Learning
- Blog: ExGRPO: Learning to Reason from Experience
- Blog: Improving Data Efficiency for LLM Reinforcement Fine-tuning Through Difficulty-targeted Online Data Selection and Rollout Replay
- Blog: Improving Data Efficiency for LLM Reinforcement Fine-tuning Through Difficulty-targeted Online Data Selection and Rollout Replay
- Blog: Knapsack RL: Unlocking Exploration of LLMs via Optimizing Budget Allocation
- Blog: Prioritized Replay for RL Post-training
- Blog: GenEnv: Difficulty-Aligned Co-Evolution Between LLM Agents and Environment Simulators
- Blog: ExGRPO: Learning to Reason from Experience
- Blog: The Eighty Five Percent Rule for optimal learning
- Blog: Can Prompt Difficulty be Online Predicted for Accelerating RL Finetuning of Reasoning Models?
- Blog: SPELL: Self-Play Reinforcement Learning for evolving Long-Context Language Models
- Blog: CurES: From Gradient Analysis to Efficient Curriculum Learning for Reasoning LLMs
- Blog: Learning on the Job: Test-Time Curricula for Targeted Reinforcement Learning
- Blog: From Blind Spots to Gains: Diagnostic-Driven Iterative Training for Large Multimodal Models
- Blog: SwS: Self-aware Weakness-driven Problem Synthesis in Reinforcement Learning for LLM Reasoning
- Blog: Exploration by Exploitation: Curriculum Learning for Reinforcement Learning Agents through Competence-Based Curriculum Policy Search
- Blog: Behavior Injection: Preparing Language Models for Reinforcement Learning
- Blog: Proximal Curriculum for Reinforcement Learning Agents
- Blog: D²Evo: Dual Difficulty-Aware Self-Evolution for Data-Efficient Reinforcement Learning
- Blog: Training Reinforcement Learning Agents and Humans With Difficulty-Conditioned Generators
- Blog: R1-Code-Interpreter: LLMs Reason with Code via Supervised and Multi-stage Reinforcement Learning
- Blog: OpenSIR: Open-Ended Self-Improving Reasoner
- Blog: Understanding Data Influence in Reinforcement Finetuning
- Blog: A Competence-Aware Curriculum for Visual Concepts Learning via Question Answering
- Blog: On the Interplay of Pre-Training, Mid-Training, and RL on Reasoning Language Models
- Blog: No Regrets: Investigating and Improving Regret Approximations for Curriculum Discovery
- Blog: Tool-R1: Sample-Efficient Reinforcement Learning for Agentic Tool Use
- Blog: Automatic Curriculum Learning For Deep RL: A Short Survey
- Blog: UFO-RL: Uncertainty-Focused Optimization for Efficient Reinforcement Learning Data Selection
- Blog: Code-driven Number Sequence Calculation: Enhancing the inductive Reasoning Abilities of Large Language Models
- Blog: D2Evo: Dual Difficulty-Aware Self-Evolution for Data-Efficient Reinforcement Learning
- Blog: UFO-RL: Uncertainty-Focused Optimization for Efficient Reinforcement Learning Data Selection
- Blog: PROS: Towards Compute-Efficient RLVR via Rollout Prefix Reuse
- Blog: Learning From Mistakes Makes LLM Better Reasoner
- Blog: HS-STAR: Hierarchical Sampling for Self-Taught Reasoners via Difficulty Estimation and Budget Reallocation
- Blog: BOTS: A Unified Framework for Bayesian Online Task Selection in LLM Reinforcement Finetuning
- Blog: Adaptive algorithms for shaping behavior
- Blog: HeaPA: Difficulty-Aware Heap Sampling and On-Policy Query Augmentation for LLM Reinforcement Learning
- Blog: A Task-centric Theory for Iterative Self-Improvement with Easy-to-Hard Curricula
- Blog: Motif-2-12.7B-Reasoning: A Practitioner's Guide to RL Training Recipes
- Blog: BOTS: A Unified Framework for Bayesian Online Task Selection in LLM Reinforcement Finetuning
- Blog: Agentic Proposing: Enhancing Large language Model Reasoning via Compositional Skill Synthesis
- Blog: PyVision-RL: Forging Open Agentic Vision Models via RL
- Blog: M2-Reasoning: Empowering MLLMs with Unified General and Spatial Reasoning
- Blog: ADORA: Training Reasoning Models with Dynamic Advantage Estimation on Reinforcement Learning
- Blog: CurES: From Gradient Analysis to Efficient Curriculum Learning for Reasoning LLMs
- Blog: Tilting the Odds at the Lottery: the Interplay of Overparameterisation and Curricula in Neural Networks
- Blog: Towards High Data Efficiency in Reinforcement Learning with Verifiable Reward
- Blog: Prompt Curriculum Learning for Efficient LLM Post-Training
- Blog: Self-Evolving Curriculum for LLM Reasoning
- Blog: Proximal Curriculum for Reinforcement Learning Agents
- Blog: Teacher algorithms for curriculum learning of Deep RL in continuously parameterized environments
- Blog: Step-Audio-R1 Technical Report
- Blog: SeRL: Self-play Reinforcement Learning for Large Language Models with Limited Data
Showing 60 of 71 papers.
History & citation
Current · Permanent link
Cite this direction
Latest version (a living page):
@misc{lacuna_intermediate_difficulty_optimization_1499964d,
title = {intermediate difficulty optimization},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/intermediate-difficulty-optimization/rdr_1499964d488641d5b8edf638a423d9ee},
note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-19},
}This exact version (version 2, frozen text):
@misc{lacuna_intermediate_difficulty_optimization_1499964d_v2,
title = {intermediate difficulty optimization},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/intermediate-difficulty-optimization/rdr_1499964d488641d5b8edf638a423d9ee/v/2},
note = {Version 2, frozen 2026-10-04. Accessed 2026-11-19},
}