6arXiv cs.CL (Computation and Language)·
Role-factorized hierarchical search agents: delegation capacity matters more than execution
A new arXiv paper investigates how model capacity should be distributed across roles in multi-agent search systems, factorizing hierarchical search into delegation, execution, and answer generation roles. Controlled sweeps across five multi-hop QA benchmarks find that scaling the delegation backbone improves exact match by ~11 points while scaling execution sub-agents yields only ~2.6 points, identifying task decomposition as the primary bottleneck. A 1.7B-parameter executor trained via trajectory distillation matches frontier sub-agent accuracy while using 37% fewer tokens, advancing the efficiency Pareto frontier. The results offer a concrete design recipe: concentrate capacity at delegation and downsize execution.
Related guides (3)
Related events (8)
Agents-A1: 35B MoE agent matches trillion-parameter models via horizon scaling
Researchers introduce Agents-A1, a 35B Mixture-of-Experts model that claims to match or exceed trillion-parameter models like Kimi-K2 and DeepSeek V4 on long-horizon agentic benchmarks. The approach scales agent trajectory length (averaging 45K tokens) and heterogeneous agent abilities rather than raw parameter count, using a three-stage training recipe including multi-teacher domain-routed distillation. On benchmarks such as SEAL-0, IFBench, HiPhO, and FrontierScience-Olympiad, Agents-A1 achieves leading or competitive results against models with roughly 30x more parameters. The work proposes a practical efficiency path for agentic capability scaling without proportional compute scaling.
AARRI-Bench evaluates frontier LLMs and agents on granular research-intern-level tasks
Researchers introduce AARR (Act As a Real Researcher), a new benchmark series targeting whether AI agents can emulate the professionalism, thoroughness, and nuanced judgment of human researchers in granular research scenarios—not just macro-level task execution. The first benchmark, AARRI-Bench, tests frontier models and agentic harnesses, finding that even the best configuration (Mini-SWE-Agent with Claude Opus 4.7) achieves only 68.3% success, frequently missing subtle but critical details obvious to human researchers. The work argues that closing the gap requires deeper modeling of research behavior rather than more complex scaffolding.
AXPO: Agent Explorative Policy Optimization Addresses Thinking-Acting Gap in Multimodal Agentic Reasoning
This paper identifies a structural asymmetry in agentic reasoning called the 'Thinking-Acting Gap,' where tool use is attempted in only ~30% of rollouts under standard RL training (GRPO), and all-wrong tool-using subgroups suppress learning signals. The authors propose AXPO (Agent eXplorative Policy Optimization), which fixes the thinking prefix and resamples tool calls for all-wrong subgroups, combined with uncertainty-based prefix selection. Evaluated across nine multimodal benchmarks on Qwen3-VL-Thinking at multiple scales, SFT+AXPO outperforms SFT+GRPO by +1.8pp on both Pass@1 and Pass@4 at 8B, with the 8B model surpassing the 32B baseline on Pass@4 using 4× fewer parameters.
Benchmark and study of LLM agents in deliberative multi-agent collaboration under partial observability
Researchers introduce a scalable benchmark for evaluating LLM agents on cooperative joint decision-making tasks where agents must exchange information under partial and asymmetric observations to reach a shared decision. A systematic evaluation of representative LLMs finds that state-of-the-art models still struggle with complex deliberative collaboration, failing in either information alignment or downstream reasoning even with external mathematical tools. Diagnostic analysis also reveals that deliberation can enable reflection and error correction, sometimes outperforming centralized baselines, offering a nuanced picture of multi-agent LLM capabilities.
PlanBench-XL: Benchmark for LLM Agent Planning in Large-Scale Tool Ecosystems
Researchers introduce PlanBench-XL, an interactive benchmark of 327 retail tasks spanning 1,665 tools designed to evaluate LLM agents on long-horizon planning under retrieval-limited tool visibility. The benchmark includes a blocking mechanism simulating real-world disruptions such as missing or failing tools, forcing agents to detect and recover from broken execution paths. Experiments on ten leading LLMs reveal severe performance degradation: GPT-5.4 drops from 51.90% accuracy in unblocked settings to 11.36% under the most severe blocking condition, highlighting fragility in adaptive planning for large, imperfect tool environments.
Role-Agent: Bootstrapping LLM Agents via Dual-Role Evolution
Role-Agent is a new framework that uses a single LLM simultaneously as both agent and environment, enabling self-bootstrapped co-evolution without external environment feedback. The system has two components: World-In-Agent (WIA), which uses predicted vs. actual state alignment as a process reward, and Agent-In-World (AIW), which reshapes training data by retrieving tasks with similar failure patterns. Experiments across multiple benchmarks show an average performance gain of over 4% over strong baselines. The approach addresses key limitations in LLM agent training: inefficient feedback and static environments.
Perplexity production data shows AI agents perform 26 min of autonomous work vs 33 sec for search, cut task time 87%
A paper using production data from Perplexity's Search and Computer products quantifies how autonomous AI agents reshape knowledge work relative to conversational search. Key findings: Computer executes 26 minutes of autonomous work per session versus 33 seconds for Search, reduces task completion time from 269 to 36 minutes on matched tasks (87% time reduction, 94% cost reduction), and lowers per-query dissatisfaction by 55%. The study also finds agents shift user behavior toward higher-order tasks, cross occupational boundaries more often, and unlock work categories essentially absent from search usage.
EdgeBench finds log-sigmoid scaling laws for agent learning from real-world environments
A new arXiv preprint introduces EdgeBench, a suite of 134 real-world tasks with ultra-long horizons (12+ hours of continuous agent operation each) spanning scientific discovery, software engineering, formal mathematics, and other domains. Analyzing ~38,000 hours of agent-environment interaction, the authors report the first evidence that agent performance during environment learning follows a log-sigmoid scaling law with R²=0.998. They also find that agent learning speed roughly doubles every three months across model generations, drawing an analogy to pretraining scaling laws but for post-deployment environment learning. 51 tasks and the evaluation framework are publicly released.