A training-free, model-agnostic dashboard that exposes per-block context state (tokens, age, budget) with lossless archive/recovery improves long-horizon tool-agent performance on LOCA-Bench, BrowseComp-Plus, and GAIA.
hub Canonical reference
Scaling long-horizon llm agent via context-folding
Canonical reference. 88% of citing Pith papers cite this work as background.
30
Pith papers citing it
Background
88% of classified citations
hub tools
citation-role summary
background 7
other 1
citation-polarity summary
representative citing papers
DeLM decentralizes LLM multi-agent coordination with shared verified context, delivering up to 10.5pp gains on SWE-bench Verified and 5.7pp on LongBench-v2 while cutting cost per task by ~50%.
VitaBench 2.0 introduces a benchmark for long-term personalized and proactive agent behavior, with results indicating substantial gaps in current frontier LLMs.
SeCo performs semantic-driven context compression for LLMs by anchoring on query-relevant semantic centers and applying consistency-weighted token merging, yielding better downstream performance, lower latency, and stronger out-of-domain robustness than position-based methods across 14 benchmarks.
ReCodeAgent uses a multi-agent system to translate and validate large code repositories across multiple programming languages, achieving 60.8% higher test pass rates than prior neuro-symbolic and agentic methods on 118 real-world projects.
Agents-A1, a 35B MoE agent, matches or exceeds selected 1T models on long-horizon agent benchmarks by scaling trajectory length and multi-domain distillation rather than parameters.
A separate memory agent that selectively injects reminders into an unmodified action agent improves long-horizon task performance by up to 8.3 percentage points.
KV-cache eviction, prompt compression, recurrent state bounding, and agent memory consolidation are unified as one rate-distortion problem with a shared lower bound, shared failure mode, and transferable mechanisms.
HORMA builds a hierarchical memory structure from agent experiences and trains a lightweight RL navigator to retrieve minimal sufficient context, yielding better task performance with at most 22.17% of baseline token usage on ALFWorld, LoCoMo, and LongMemEval.
TMEM lets LLM agents evolve their policy mid-episode by absorbing distilled supervision into online LoRA updates, outperforming summary and retrieval baselines on several long-context benchmarks.
EvoDS adds autonomous skill acquisition via synthesis-validation-reuse and adaptive context compression via learned control within a two-stage multi-agent RL scheme, claiming 28.9% average gains over prior agents on four benchmarks plus elimination of out-of-token failures.
An economy of agents using auctions and wealth accumulation produces emergent multi-step reasoning that outperforms monolithic baselines on five agentic tasks.
AgentFugue introduces a plug-in shared reasoning hub trained with SFT and RL that enables peer agents to share intermediate reasoning, yielding gains on long-horizon tasks over strong baselines.
SAM is a standalone memory framework for long-horizon LLM agents that creates state-adaptive cues from interactions, preserves raw trajectories for intent-driven recall, and optimizes the module via expert supervision and RL, outperforming baselines on BrowseComp and related benchmarks.
PEEK maintains a constant-sized context map via a programmable cache policy to give LLM agents persistent orientation knowledge about recurring external contexts, yielding 6-34% gains and lower cost than prior prompt-learning methods.
Position-preserving MASK token compression reduces redundancy in diffusion LLMs to accelerate parallel decoding and enable context folding for longer sequences.
PruneTIR prunes erroneous tool-call trajectories during LLM inference via three trigger-based components to raise Pass@1 accuracy and efficiency while shortening context.
Agentic memory improves clean reasoning but worsens performance when spurious patterns are present in stored trajectories; CAMEL calibration reduces this reliance while preserving clean performance.
Slipstream uses asynchronous compaction with trajectory-grounded judge validation to improve long-horizon agent accuracy by up to 8.8 percentage points and reduce latency by up to 39.7%.
Terminus-4B, a Qwen3-4B post-trained with SFT and rubric-based GRPO, matches or beats frontier models as a terminal-execution subagent while cutting main-agent tokens ~30% on SWE-Bench-style tasks.
LightThinker++ adds explicit adaptive memory management and a trajectory synthesis pipeline to LLM reasoning, cutting peak token use by ~70% while gaining accuracy in standard and long-horizon agent tasks.
Survey that defines agentic RL for LLMs via POMDPs, introduces a taxonomy of planning/tool-use/memory/reasoning capabilities and domains, and compiles open environments from over 500 papers.
CompactionRL trains LLM agents to generate context summaries during RL rollouts, enabling long-horizon task completion under fixed context budgets with consistent gains on SWE-bench Verified and Terminal-Bench 2.0.
SWE-MeM introduces adaptive memory management for coding agents via synthesized trajectories and Memory-aware GRPO, reporting 43.4% and 60.2% resolve rates on SWE-Bench Verified for 4B and 30B models while beating baselines on performance and token use.
citing papers explorer
-
LLM Agents Are Latent Context Managers: Eliciting Self-Managed Context via State Proprioception
A training-free, model-agnostic dashboard that exposes per-block context state (tokens, age, budget) with lossless archive/recovery improves long-horizon tool-agent performance on LOCA-Bench, BrowseComp-Plus, and GAIA.
-
Decentralized Multi-Agent Systems with Shared Context
DeLM decentralizes LLM multi-agent coordination with shared verified context, delivering up to 10.5pp gains on SWE-bench Verified and 5.7pp on LongBench-v2 while cutting cost per task by ~50%.
-
VitaBench 2.0: Evaluating Personalized and Proactive Agents in Long-Term User Interactions
VitaBench 2.0 introduces a benchmark for long-term personalized and proactive agent behavior, with results indicating substantial gaps in current frontier LLMs.
-
Beyond Position Bias: Shifting Context Compression from Position-Driven to Semantic-Driven
SeCo performs semantic-driven context compression for LLMs by anchoring on query-relevant semantic centers and applying consistency-weighted token merging, yielding better downstream performance, lower latency, and stronger out-of-domain robustness than position-based methods across 14 benchmarks.
-
ReCodeAgent: A Multi-Agent Workflow for Language-agnostic Translation and Validation of Large-scale Repositories
ReCodeAgent uses a multi-agent system to translate and validate large code repositories across multiple programming languages, achieving 60.8% higher test pass rates than prior neuro-symbolic and agentic methods on 118 real-world projects.
-
Scaling the Horizon, Not the Parameters: Reaching Trillion-Parameter Performance with a 35B Agent
Agents-A1, a 35B MoE agent, matches or exceeds selected 1T models on long-horizon agent benchmarks by scaling trajectory length and multi-domain distillation rather than parameters.
-
Remember When It Matters: Proactive Memory Agent for Long-Horizon Agents
A separate memory agent that selectively injects reminders into an unmodified action agent improves long-horizon task performance by up to 8.3 percentage points.
-
What to Keep, What to Forget: A Rate--Distortion View of Memory Compaction in LLMs and Agents
KV-cache eviction, prompt compression, recurrent state bounding, and agent memory consolidation are unified as one rate-distortion problem with a shared lower bound, shared failure mode, and transferable mechanisms.
-
Organize then Retrieve: Hierarchical Memory Navigation for Efficient Agents
HORMA builds a hierarchical memory structure from agent experiences and trains a lightweight RL navigator to retrieve minimal sufficient context, yielding better task performance with at most 22.17% of baseline token usage on ALFWorld, LoCoMo, and LongMemEval.
-
Scaling Self-Evolving Agents via Parametric Memory
TMEM lets LLM agents evolve their policy mid-episode by absorbing distilled supervision into online LoRA updates, outperforming summary and retrieval baselines on several long-context benchmarks.
-
EvoDS: Self-Evolving Autonomous Data Science Agent with Skill Learning and Context Management
EvoDS adds autonomous skill acquisition via synthesis-validation-reuse and adaptive context compression via learned control within a two-stage multi-agent RL scheme, claiming 28.9% average gains over prior agents on four benchmarks plus elimination of out-of-token failures.
-
Economy of Minds: Emerging Multi-Agent Intelligence with Economic Interactions
An economy of agents using auctions and wealth accumulation produces emergent multi-step reasoning that outperforms monolithic baselines on five agentic tasks.
-
AgentFugue: Agent Scaling for Long-Horizon Tasks through Collective Reasoning
AgentFugue introduces a plug-in shared reasoning hub trained with SFT and RL that enables peer agents to share intermediate reasoning, yielding gains on long-horizon tasks over strong baselines.
-
SAM: State-Adaptive Memory for Long-Horizon Reasoning Agent
SAM is a standalone memory framework for long-horizon LLM agents that creates state-adaptive cues from interactions, preserves raw trajectories for intent-driven recall, and optimizes the module via expert supervision and RL, outperforming baselines on BrowseComp and related benchmarks.
-
PEEK: Context Map as an Orientation Cache for Long-Context LLM Agents
PEEK maintains a constant-sized context map via a programmable cache policy to give LLM agents persistent orientation knowledge about recurring external contexts, yielding 6-34% gains and lower cost than prior prompt-learning methods.
-
Elastic-dLLM: Position Preserving Context Compression and Augmentation of Diffusion LLMs
Position-preserving MASK token compression reduces redundancy in diffusion LLMs to accelerate parallel decoding and enable context folding for longer sequences.
-
PruneTIR: Inference-Time Tool Call Pruning for Effective yet Efficient Tool-Integrated Reasoning
PruneTIR prunes erroneous tool-call trajectories during LLM inference via three trigger-based components to raise Pass@1 accuracy and efficiency while shortening context.
-
The Trap of Trajectory: Towards Understanding and Mitigating Spurious Correlations in Agentic Memory
Agentic memory improves clean reasoning but worsens performance when spurious patterns are present in stored trajectories; CAMEL calibration reduces this reliance while preserving clean performance.
-
Slipstream: Trajectory-Grounded Compaction Validation for Long-Horizon Agents
Slipstream uses asynchronous compaction with trajectory-grounded judge validation to improve long-horizon agent accuracy by up to 8.8 percentage points and reduce latency by up to 39.7%.
-
Terminus-4B: Can a Smaller Model Replace Frontier LLMs at Agentic Execution Tasks?
Terminus-4B, a Qwen3-4B post-trained with SFT and rubric-based GRPO, matches or beats frontier models as a terminal-execution subagent while cutting main-agent tokens ~30% on SWE-Bench-style tasks.
-
LightThinker++: From Reasoning Compression to Memory Management
LightThinker++ adds explicit adaptive memory management and a trajectory synthesis pipeline to LLM reasoning, cutting peak token use by ~70% while gaining accuracy in standard and long-horizon agent tasks.
-
The Landscape of Agentic Reinforcement Learning for LLMs: A Survey
Survey that defines agentic RL for LLMs via POMDPs, introduces a taxonomy of planning/tool-use/memory/reasoning capabilities and domains, and compiles open environments from over 500 papers.
-
CompactionRL: Reinforcement Learning with Context Compaction for Long-Horizon Agents
CompactionRL trains LLM agents to generate context summaries during RL rollouts, enabling long-horizon task completion under fixed context budgets with consistent gains on SWE-bench Verified and Terminal-Bench 2.0.
-
SWE-MeM: Learning Adaptive Memory Management for Long-Horizon Coding Agents
SWE-MeM introduces adaptive memory management for coding agents via synthesized trajectories and Memory-aware GRPO, reporting 43.4% and 60.2% resolve rates on SWE-Bench Verified for 4B and 30B models while beating baselines on performance and token use.
-
ActiveMem: Distributed Active Memory for Long-Horizon LLM Reasoning
ActiveMem proposes a heterogeneous distributed memory framework for LLM agents that separates planning from active memory management, reporting SOTA accuracy with lower overhead on BrowseComp-Plus and GAIA.
-
Code as Agent Harness
A survey that organizes existing work on LLM-based agents around code as the central harness, structured in three layers of interfaces, mechanisms, and multi-agent scaling, with applications across domains and listed open challenges.
-
Context Pruning for Coding Agents via Multi-Rubric Latent Reasoning
LaMR decomposes code context pruning into two rubrics using dedicated CRFs, a mixture-of-experts gate, and AST-derived labels to filter noise and often match or beat full-context baselines on coding benchmarks.
-
On Training Large Language Models for Long-Horizon Tasks: An Empirical Study of Horizon Length
Longer action horizons bottleneck LLM agent training through instability, but training with reduced horizons stabilizes learning and enables better generalization to longer horizons.
-
HIPIF: Hierarchical Planning and Information Folding for Long-Horizon LLM Agent Learning
HIPIF trains LLM agents end-to-end using subgoal-based hierarchical planning and information folding of completed histories, plus hierarchical reflection and process rewards, to handle long-horizon tasks without auxiliary models or expert trajectories.
-
Reinforcement Learning for LLM-based Multi-Agent Systems through Orchestration Traces
This survey organizes RL for LLM multi-agent systems into reward families, credit units, and five orchestration sub-decisions, notes the absence of explicit stopping-decision training in its paper pool, and releases a tagged corpus.