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.
Reasoning aware self-consistency: Lever- aging reasoning paths for efficient llm sampling
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
citation-role summary
background 2
citation-polarity summary
roles
background 2polarities
background 2representative citing papers
LRMs underperform on simple system 1 questions in both accuracy and efficiency, with problem difficulty implicitly encoded in early hidden states.
A survey organizing techniques to achieve efficient reasoning in LLMs by shortening chain-of-thought outputs.
Empirical evaluation on Gemini 2.5 models shows self-consistency yields only 0.4% gain on HotpotQA and 1.6% on MATH-500 across 20 samples while token costs scale linearly, with performance plateauing or declining at higher counts.
A survey that deconstructs LLM agent systems via a methodology-centered taxonomy linking design principles to emergent behaviors, applications, and challenges.
citing papers explorer
-
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.
-
Exploring the System 1 Thinking Capability of Large Reasoning Models
LRMs underperform on simple system 1 questions in both accuracy and efficiency, with problem difficulty implicitly encoded in early hidden states.
-
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.
-
Self-Consistency Is Losing Its Edge: Diminishing Returns and Rising Costs in Modern LLMs
Empirical evaluation on Gemini 2.5 models shows self-consistency yields only 0.4% gain on HotpotQA and 1.6% on MATH-500 across 20 samples while token costs scale linearly, with performance plateauing or declining at higher counts.
-
Large Language Model Agent: A Survey on Methodology, Applications and Challenges
A survey that deconstructs LLM agent systems via a methodology-centered taxonomy linking design principles to emergent behaviors, applications, and challenges.