Papers arxiv:2605.08083 LLMs Improving LLMs: Agentic Discovery for Test-Time Scaling Published on May 8 Submitted by Chengsong Huang on May 11 3 Paper of the day Google Authors: Tong Zheng , , , , , , Runpeng Dai , , , Tianyi Xiong , , , Abstract AutoTTS automates test-time scaling strategy discove… …
… The following papers were recommended by the Semantic Scholar API From History to State: Constant-Context Skill Learning for LLM Agents 2026 LongMemEval-V2: Evaluating Long-Term Agent Memory Toward Experienced Colleagues 2026 ClawGym: A Scalable Framework for Building Effective Claw Agents 2026 Ali… …
… The following papers were recommended by the Semantic Scholar API PlanCompiler: A Deterministic Compilation Architecture for Structured Multi-Step LLM Pipelines 2026 Complete Cyclic Subtask Graphs for Tool-Using LLM Agents: Flexibility, Cost, and Bottlenecks in Multi-Agent Workflows 2026 ClawForge:… …
… Notably, if you are worried about the high cost of deep research, Pi-Serini may help: it reduces evaluation cost by 3.3×–10×. …
… The following papers were recommended by the Semantic Scholar API PRAISE: Prefix-Based Rollout Reuse in Agentic Search Training 2026 Train at Moving Edge: Online-Verified Prompt Selection for Efficient RL Training of Large Reasoning Model 2026 Accelerating RL Post-Training Rollouts via System-Integ… …
… AI-generated summary LLM-based evolution has emerged as a promising way to improve agents by refining non-parametric artifacts, but its wall-clock cost remains a major bottleneck. We identify that this cost comes from synchronized stage execution and imbalance inside each LLM-heavy stage. …
… The following papers were recommended by the Semantic Scholar API Uno-Orchestra: Parsimonious Agent Routing via Selective Delegation 2026 EvoMAS: Learning Execution-Time Workflows for Multi-Agent Systems 2026 From Model Scaling to System Scaling: Scaling the Harness in Agentic AI 2026 Coordination … …
… Both provide limited control: Prompt-only suppresses necessary calls alongside unnecessary ones, and Reason-then-Act still incurs a disproportionate accuracy cost on hard tasks. …
… We propose vOPD On-Policy Distillation with a control variate baseline , which casts OPD as policy-gradient RL and stabilizes it by introducing a control variate baseline-canonically a value function -- from the RL literature. …
… Existing safety-alignment signals are largely response-level or off-policy, and often incur a safety-utility trade-off: improving agent safety comes at the cost of degraded task performance . …