Fork-think with confidence identifies forking points via model confidence in a single path before sampling continuations, cutting tokens up to 30% and runtime up to 57% on reasoning benchmarks while matching or exceeding parallel thinking performance.
Token-budget-aware LLM reasoning
8 Pith papers cite this work, alongside 10 external citations. Polarity classification is still indexing.
8
Pith papers citing it
10
external citations · OpenAlex
years
2026 8representative citing papers
Under matched lost-correct risk, learned early exits beat calibrated scalar exits on free-form math, lose on multiple-choice, and neither certifies savings on small hard sets; serving regime can reverse the savings.
Dynamic Rollout Editing reduces overthinking in RL-trained LLMs by editing post-answer continuations in successful rollouts and preferring the edited versions within GRPO groups.
Post-hoc model-based compression of reasoning traces cuts training tokens to 12-30% and speeds training 2-7.6x while retaining up to 96% of raw-trace accuracy, though raw traces remain superior at every scale.
PUMA detects reasoning-level semantic redundancy to enable early exit in chains of thought, achieving 26.2% average token reduction across five LRMs and five benchmarks while preserving accuracy and CoT quality.
SlimSearcher reduces tool-call rounds by 17-58% on GAIA, BrowseComp and XBenchDeepSearch while maintaining accuracy via Pareto filtration in SFT and Adaptive Reward Gating in RL.
SAT reduces reasoning tokens by up to 40% across multiple large reasoning models and benchmarks by adaptively pruning steps based on difficulty while maintaining or improving accuracy.
Apriel-1.5-OpenReasoner uses RL post-training with adaptive sampling and difficulty-aware penalties to boost reasoning accuracy on AIME, GPQA, MMLU-Pro and LiveCodeBench while producing shorter traces and generalizing beyond its 16K training budget.
citing papers explorer
-
Fork-Think with Confidence
Fork-think with confidence identifies forking points via model confidence in a single path before sampling continuations, cutting tokens up to 30% and runtime up to 57% on reasoning benchmarks while matching or exceeding parallel thinking performance.
-
When Does Learning to Stop Help? A Cost-Aware Study of Early Exits in Reasoning Models
Under matched lost-correct risk, learned early exits beat calibrated scalar exits on free-form math, lose on multiple-choice, and neither certifies savings on small hard sets; serving regime can reverse the savings.
-
Dynamic Rollout Editing for Reducing Overthinking in RL-Trained Reasoning Models
Dynamic Rollout Editing reduces overthinking in RL-trained LLMs by editing post-answer continuations in successful rollouts and preferring the edited versions within GRPO groups.
-
Compress-Distill: Reasoning Trace Compression for Efficient Knowledge Distillation
Post-hoc model-based compression of reasoning traces cuts training tokens to 12-30% and speeds training 2-7.6x while retaining up to 96% of raw-trace accuracy, though raw traces remain superior at every scale.
-
Stop When Reasoning Converges: Semantic-Preserving Early Exit for Reasoning Models
PUMA detects reasoning-level semantic redundancy to enable early exit in chains of thought, achieving 26.2% average token reduction across five LRMs and five benchmarks while preserving accuracy and CoT quality.
-
SlimSearcher: Training Efficiency-Aware Web Agents via Adaptive Reward Gating
SlimSearcher reduces tool-call rounds by 17-58% on GAIA, BrowseComp and XBenchDeepSearch while maintaining accuracy via Pareto filtration in SFT and Adaptive Reward Gating in RL.
-
SAT: Balancing Reasoning Accuracy and Efficiency with Stepwise Adaptive Thinking
SAT reduces reasoning tokens by up to 40% across multiple large reasoning models and benchmarks by adaptively pruning steps based on difficulty while maintaining or improving accuracy.
-
Apriel-1.5-OpenReasoner: RL Post-Training for General-Purpose and Efficient Reasoning
Apriel-1.5-OpenReasoner uses RL post-training with adaptive sampling and difficulty-aware penalties to boost reasoning accuracy on AIME, GPQA, MMLU-Pro and LiveCodeBench while producing shorter traces and generalizing beyond its 16K training budget.