9 papers
AbridgeKV: Compressing Long Reasoning into the KV Cache of Learned Pause Tokens
Jack Cai, Lijie Yang, Vasilis Kontonis +2
Long chain-of-thought decoding is memory-bound: the KV cache of the generated tokens grows linearly with the test-time compute being spent. Evicting cache entries does not fix this…
Fork-and-Flush: Escaping Idea Basins in Autoresearch Agents
Ziyang Cai, Christos Ziakas, Vasilis Kontonis +5
Autoresearch agents tackle open-ended problems by repeatedly proposing candidate solutions, evaluating them, and using feedback to guide subsequent experiments. We show that indepe…
Scaling Discovery through Test-Time Communication
Jongho Park, Vasilis Kontonis, Shivam Garg +2
Science advances not in isolation but through collaboration, yet existing agentic systems capture little of this. Whether communicating agents help remains an open question with mi…
SuperThoughts: Reasoning Tokens in Superposition
Zheyang Xiong, Shivam Garg, Max Yu +4
Long Chain-of-Thought (CoT) reasoning improves LLM problem-solving but is computationally expensive due to sequential token generation. While recent works explore reasoning in cont…
MEMENTO: Teaching LLMs to Manage Their Own Context
Vasilis Kontonis, Yuchen Zeng, Shivam Garg +7
Reasoning models think in long, unstructured streams with no mechanism for compressing or organizing their own intermediate state. We introduce MEMENTO: a method that teaches model…
Endless Terminals: Scaling RL Environments for Terminal Agents
Kanishk Gandhi, Shivam Garg, Noah D. Goodman +1
Environments are the bottleneck for self-improving agents. Current terminal benchmarks were built for evaluation, not training; reinforcement learning requires a scalable pipeline,…
Wait, Wait, Wait... Why Do Reasoning Models Loop?
Charilaos Pipis, Shivam Garg, Vasilis Kontonis +3
Reasoning models (e.g., DeepSeek-R1) generate long chains of thought to solve harder problems, but they often loop, repeating the same text at low temperatures or with greedy decod…
Sample More to Think Less: Group Filtered Policy Optimization for Concise Reasoning
Vaishnavi Shrivastava, Ahmed Awadallah, Vidhisha Balachandran +3
Large language models trained with reinforcement learning with verifiable rewards tend to trade accuracy for length--inflating response lengths to achieve gains in accuracy. While…
Inference-Time Scaling for Complex Tasks: Where We Stand and What Lies Ahead
Vidhisha Balachandran, Jingya Chen, Lingjiao Chen +8
Inference-time scaling can enhance the reasoning capabilities of large language models (LLMs) on complex problems that benefit from step-by-step problem solving. Although lengtheni…