AIBullisharXiv – CS AI · Jun 237/10
🧠MemoryVAM introduces an episodic memory mechanism for video-world-model policies that enables robots to perform long-horizon manipulation tasks by retaining and leveraging historical context. The system achieves significant performance improvements on benchmark tasks and real robot experiments, addressing a fundamental limitation where short observation windows make complex manipulation non-Markovian.
AIBullisharXiv – CS AI · Jun 237/10
🧠Researchers propose a hierarchical reinforcement learning method that combines learned world models with dual-level policies to enable safe exploration in long-horizon tasks. The approach uses high-level subgoals to guide exploration toward safe regions and low-level imagined rollouts to minimize unsafe behaviors, demonstrating significant improvements over existing Safe RL baselines on complex navigation and manipulation tasks.
AIBearisharXiv – CS AI · Jun 237/10
🧠Researchers demonstrate that large language model agents fail to maintain plans as persistent internal state, instead relying on plans remaining in the context window. Using diagnostic techniques on Llama-3.1-70B and DeepSeek-R1, the study shows plan signal decays rapidly when compressed out of context, with practical implications for agent reliability in long-horizon tasks.
🧠 Llama
AIBullisharXiv – CS AI · Jun 197/10
🧠Researchers present the 'Connect the Dots' (CoD) framework for training large language models to function as long-lifecycle agents that learn from experience and progressively improve performance across tasks. The work combines reinforcement learning with self-updating context mechanisms, demonstrating cross-domain generalization capabilities and releasing implementations to advance AI agent research.
AIBullisharXiv – CS AI · Jun 107/10
🧠Researchers introduce ActiveMem, a distributed memory framework that decouples storage from reasoning in large language models, enabling agents to handle longer tasks without context overload. The system separates executive planning from memory management—inspired by human brain architecture—and demonstrates state-of-the-art performance on complex reasoning benchmarks while reducing computational overhead.
AINeutralarXiv – CS AI · Jun 97/10
🧠Researchers introduce SWE-Marathon, a benchmark testing AI agents on 20 ultra-long-horizon software engineering tasks requiring millions of tokens and hours of sustained work. Current frontier coding agents solve fewer than 30% of tasks, revealing critical gaps in planning, self-verification, and memory management that limit real-world deployment.
AINeutralarXiv – CS AI · Jun 97/10
🧠Researchers introduce WeaveBench, a comprehensive benchmark for evaluating computer-use agents across hybrid interfaces combining GUI, CLI, and code operations. The benchmark reveals significant capability gaps, with the best frontier models achieving only 41.2% success rates on 114 real-world tasks, indicating that current AI agents struggle with complex multi-interface orchestration.
AIBullisharXiv – CS AI · Jun 57/10
🧠Researchers introduce MAGE, a novel memory management system for LLM-based agents that organizes task histories as hierarchical state trees rather than semantic similarity clusters. The approach achieves 7.8-20.4 percentage point improvements in task success rates while reducing token consumption by 55.1% on long-horizon tasks with interdependent decisions.
AIBullisharXiv – CS AI · Jun 27/10
🧠Researchers introduce ACON, a framework that compresses long-context information for LLM agents without model fine-tuning, reducing token usage by 26-54% while improving task success rates. The method optimizes compression through natural language refinement and enables smaller language models to function effectively as long-horizon agents.
AIBullisharXiv – CS AI · Jun 17/10
🧠Researchers introduce ASH, an agentic system that learns embodied policies from unlabeled internet video without reward shaping or expert demonstration. Through a self-improvement loop using Inverse Dynamics Models, ASH achieves sustained progression on long-horizon tasks in Pokemon Emerald and Legend of Zelda, significantly outperforming baseline approaches.
AIBullisharXiv – CS AI · May 287/10
🧠Researchers propose Periodic RoPE (P-RoPE), a novel positional encoding mechanism that combines sliding window attention for local dependencies with global attention layers lacking positional constraints, enabling language models to theoretically support infinite context windows without performance degradation. The approach addresses a fundamental limitation in current LLMs where model performance degrades when sequence length exceeds the pre-trained range of positional encodings like RoPE.
AIBearishDecrypt – AI · May 277/10
🧠Huawei has introduced Claw-Anything, a benchmark that tests AI agents' ability to handle complex digital tasks over extended simulated timeframes. GPT-5.5, currently the best-performing model, achieved only 34.5% on the benchmark, highlighting significant limitations in current AI agents' capacity to maintain performance during long-horizon tasks.
🧠 GPT-5
AIBullisharXiv – CS AI · May 127/10
🧠Researchers propose DeMem, a decision-centric memory framework that optimizes agent memory allocation based on preserving distinctions needed for sound decision-making rather than descriptive accuracy. Using rate-distortion theory, the approach identifies what information can be safely forgotten under memory constraints and demonstrates performance gains on long-horizon language agent tasks.
AIBullisharXiv – CS AI · May 127/10
🧠Researchers propose Agent Cybernetics, a theoretical framework applying mid-20th century control systems theory to modern LLM-based AI agents. The framework addresses critical gaps in how foundation agents are designed, offering scientific principles for reliability, continuous operation, and safe self-improvement across long-horizon tasks.
AIBullisharXiv – CS AI · May 97/10
🧠ReFlect introduces a training-free harness system that wraps around LLMs to detect and recover from reasoning failures in complex, multi-step tasks. Testing across six models shows significant improvements in task success rates, with gains inversely correlated to baseline performance, though the approach reveals limitations in how smaller models handle structured reasoning.
🧠 GPT-4🧠 Claude🧠 Sonnet
AIBullisharXiv – CS AI · May 97/10
🧠Researchers introduce BEACON, a milestone-guided policy learning framework that significantly improves training efficiency for long-horizon language agents by solving credit misattribution and sample inefficiency problems. The approach achieves 92.9% success rates on complex tasks—nearly double previous benchmarks—while improving sample utilization from 23.7% to 82.0%.
AINeutralarXiv – CS AI · Apr 157/10
🧠Researchers introduce HORIZON, a diagnostic benchmark for identifying and analyzing why large language model agents fail at long-horizon tasks requiring extended action sequences. By evaluating state-of-the-art models across multiple domains and proposing an LLM-as-a-Judge attribution pipeline, the study provides systematic methodology for understanding agent limitations and improving reliability.
🧠 GPT-5🧠 Claude
AIBullisharXiv – CS AI · Mar 167/10
🧠Research shows that large language models' performance on short tasks may underestimate their capabilities, as small improvements in single-step accuracy lead to exponential gains in handling longer tasks. The study reveals that larger models excel at execution over many steps, though they suffer from 'self-conditioning' where previous errors increase the likelihood of future mistakes, which can be mitigated through 'thinking' mechanisms.
AIBullisharXiv – CS AI · Mar 57/10
🧠Researchers developed ELMUR, a new AI architecture that uses external memory to help robots make better decisions over extremely long time periods. The system achieved 100% success on tasks requiring memory of up to one million steps and nearly doubled performance on robotic manipulation tasks compared to existing methods.
AINeutralarXiv – CS AI · Jun 236/10
🧠Researchers introduce RARM (Reference-Anchored Reward Model), a visual AI system that solves a major bottleneck in robot learning by converting single successful demonstrations into dense reward signals without task-specific engineering. The approach uses confidence-gated progress matching to avoid false-positive rewards, achieving superior performance across simulated and real-world manipulation tasks.
AIBullisharXiv – CS AI · Jun 116/10
🧠Researchers introduce PRInTS, a generative process reward model designed to improve AI agents' ability to perform multi-step information-seeking tasks over long horizons. By combining dense scoring across multiple quality dimensions with trajectory summarization, PRInTS enables smaller language models to match or exceed frontier model performance on complex reasoning benchmarks.
AINeutralarXiv – CS AI · Jun 106/10
🧠Researchers introduce OSL-MR, a framework that optimizes memory retention for long-horizon language agents by treating it as a constrained optimization problem rather than local decisions. The approach combines learned evidence valuation with heuristic scoring while respecting real-world observability constraints, demonstrating superior performance over existing methods on benchmark datasets.
AINeutralarXiv – CS AI · Jun 106/10
🧠Researchers introduce Workflow-GYM, a benchmark for evaluating AI agents on complex, long-horizon professional GUI tasks across specialized software environments. Testing reveals that even state-of-the-art models achieve only 30% success rates, exposing significant limitations in agent consistency, error handling, and domain-specific software comprehension.
AINeutralarXiv – CS AI · Jun 106/10
🧠Researchers propose HIPIF, a novel training method that improves Large Language Model agents' performance on complex multi-step tasks by organizing execution around explicit subgoals and summarizing completed progress to reduce interference from growing context. The approach combines hierarchical planning with reward mechanisms, demonstrating improvements on three public benchmarks without requiring costly auxiliary models.
AIBullisharXiv – CS AI · Jun 96/10
🧠Researchers present SearchSwarm, a framework that trains large language models to intelligently delegate complex tasks to subagents while managing finite context windows. The resulting 30B-parameter model achieves state-of-the-art performance on research benchmarks by learning when and what to delegate, addressing a critical bottleneck in agentic AI systems.
