POPO uses bounded importance sampling on positive rollouts and a siamese policy network to achieve implicit negative gradients and stable optimization, matching or exceeding GRPO on math benchmarks such as 36.67% on AIME 2025.
Inftythink: Breaking the length limits of long-context reasoning in large language models
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
representative citing papers
Post-Reasoning boosts LLM accuracy by reversing the usual answer-after-reasoning order, delivering mean relative gains of 17.37% across 117 model-benchmark pairs with zero extra cost.
ZoomR reduces KV cache memory by more than 4x during long-output reasoning by using summary keys for coarse indexing and dynamic fine-grained retrieval.
MEMENTO trains LLMs to segment reasoning into blocks, generate mementos as dense summaries, and reason forward using only mementos and KV states, cutting peak KV cache by ~2.5x while preserving benchmark accuracy.
SWE-AGILE introduces a Dynamic Reasoning Context with sliding windows of detailed steps and compressed Reasoning Digests to enable efficient long-horizon reasoning in software engineering agents, claiming new benchmark results on SWE-Bench-Verified for 7B-8B models.
A survey organizing techniques to achieve efficient reasoning in LLMs by shortening chain-of-thought outputs.
citing papers explorer
-
Beyond Negative Rollouts: Positive-Only Policy Optimization with Implicit Negative Gradients
POPO uses bounded importance sampling on positive rollouts and a siamese policy network to achieve implicit negative gradients and stable optimization, matching or exceeding GRPO on math benchmarks such as 36.67% on AIME 2025.
-
Post Reasoning: Improving the Performance of Non-Thinking Models at No Cost
Post-Reasoning boosts LLM accuracy by reversing the usual answer-after-reasoning order, delivering mean relative gains of 17.37% across 117 model-benchmark pairs with zero extra cost.
-
ZoomR: Memory Efficient Reasoning through Multi-Granularity Key Value Retrieval
ZoomR reduces KV cache memory by more than 4x during long-output reasoning by using summary keys for coarse indexing and dynamic fine-grained retrieval.
-
MEMENTO: Teaching LLMs to Manage Their Own Context
MEMENTO trains LLMs to segment reasoning into blocks, generate mementos as dense summaries, and reason forward using only mementos and KV states, cutting peak KV cache by ~2.5x while preserving benchmark accuracy.
-
SWE-AGILE: A Software Agent Framework for Efficiently Managing Dynamic Reasoning Context
SWE-AGILE introduces a Dynamic Reasoning Context with sliding windows of detailed steps and compressed Reasoning Digests to enable efficient long-horizon reasoning in software engineering agents, claiming new benchmark results on SWE-Bench-Verified for 7B-8B models.
-
Stop Overthinking: A Survey on Efficient Reasoning for Large Language Models
A survey organizing techniques to achieve efficient reasoning in LLMs by shortening chain-of-thought outputs.