KV-PRM: Efficient Process Reward Modeling via KV-Cache Transfer for Multi-Agent Test-Time Scaling
A new paper introduces KV-PRM, a process reward model that reuses KV-cache from the base model to avoid re-encoding entire trajectories, reducing scoring cost from quadratic to linear in sequence length. This enables efficient test-time scaling for long-context multi-agent systems.
Entities
Related
Two papers propose token-adaptive KV cache compression for long-context LLMs
Two arXiv papers from July 7, 2026 introduce token-adaptive KV cache compression methods for long-context LLM inference. DepthWeave-KV factorizes key/value states across neighboring layers using shared low-rank bases with token-specific residuals. FreqDepthKV uses shared low-frequency depth components and sparse high-frequency residuals, with an online probe assigning attention heads to different cache modes. Both aim to reduce memory bandwidth while preserving retrieval and reasoning quality.
Researchers propose memory agent to combat behavioral state decay in long-horizon tasks
A new arXiv paper identifies 'behavioral state decay' as a failure mode where decision-relevant information gets buried in long trajectories. The authors propose a separate memory agent that actively updates a structured memory bank alongside an unmodified action agent, rather than relying on passive retrieval.
Paper proposes correlation-aware bandits with surrogate rewards for LLM routing
A new arXiv paper introduces contextual bandit algorithms that leverage surrogate reward signals from machine learning models to improve LLM routing decisions. The approach accounts for inter-arm correlations and noisy auxiliary rewards, addressing limitations of classical bandits that assume conditional independence.
Long-Horizon-Terminal-Bench: New Benchmark Tests AI Agents on Long-Horizon Tasks with Dense Rewards
Researchers introduced Long-Horizon-Terminal-Bench, a benchmark of 46 long-horizon tasks across nine categories, including experiment reproduction and software engineering. Unlike existing benchmarks that evaluate only final outcomes, it uses dense reward signals to measure intermediate progress, providing a more complete picture of agent capability.
Paper proposes modeling forgetting as interference between tasks in continual learning
A new arXiv paper argues that forgetting in continual learning should be modeled directly as interference between tasks, rather than relying on post-hoc mechanisms like replay or regularization. The authors show that in the frozen-feature regime, forgetting equals the interference energy induced on the old task, and they recover this quantity in deep networks via path-averaged curvature with minimal extra computation.