Code that accompanies the paper release for "LLMs Corrupt Your Documents When You Delegate"
-
Updated
Apr 20, 2026 - Python
#
long-horizon
Here are 16 public repositories matching this topic...
Fixing GRPO training collapse in long-horizon multi-tool agents. A lightweight PRM-Lite + LATA joint approach achieves +37% over vanilla GRPO on τ-bench airline (50-task, multi-turn).
reinforcement-learning long-horizon qwen agentic-ai tool-calling process-reward-model grpo tau-bench multi-turn-agents
-
Updated
May 11, 2026 - Python
🧠 Awesome Memory-VLA: A curated list of Visual-Language-Action models with memory
robotics memory vla pomdp vlm embodied-ai long-horizon visual-language-models long-context-modeling visual-language-action-models memory-vlm memory-vla
-
Updated
May 3, 2026
SWE-Marathon: an ultra long-horizon SWE benchmark
-
Updated
May 16, 2026 - Rust
Code for Scalable Offline Model-Based RL with Action chunking
reinforcement-learning model-based-reinforcement-learning offline-reinforcement-learning long-horizon action-chunking
-
Updated
Feb 20, 2026 - Python
Code for Tackling Long-Horizon Tasks with Model-based Offline Reinforcement Learning
reinforcement-learning model-based-reinforcement-learning offline-reinforcement-learning long-horizon
-
Updated
Feb 6, 2025 - Python
The simplest way to build long-horizon environments
-
Updated
Apr 8, 2026 - Python
A family of long-horizon software-engineering environments for OpenEnv, adapted from https://github.com/Proximal-Labs/frontier-swe
-
Updated
Apr 26, 2026 - C
VLM-RL Hierarchical Loco-Manupilation For Long-Horizon Tasks With G1 robot in Isaac Lab/Sim
vlm g1 ppo semantic-map isaacsim loco-manipulation isaac-sim unitree long-horizon long-horizon-robotic-manipulation isaac-lab isaaclab unitree-g1 long-horizon-ai long-horizon-manipulation long-horizon-intelligence long-horizon-tasks
-
Updated
Apr 28, 2026 - Python
OpenClaw humanity infusions OtherPowers Creative Intelligence Agency. 🦞
-
Updated
Feb 24, 2026 - TypeScript
A real-world inspired environment for selective context retention under noise. It evaluates an LLM's ability to manage a fixed-capacity memory buffer, retaining high-value information while filtering out distractors
-
Updated
Apr 25, 2026 - Jupyter Notebook
Dashboard for real-world inspired environment for selective context retention under noise. It evaluates an LLM's ability to manage a fixed-capacity memory buffer, retaining high-value information while filtering out distractors
-
Updated
Apr 25, 2026 - Python
TPipe is the agent operating environment for deterministic, multimodal AI systems. Built Kotlin-first, it composes runtime substrates into governed pipelines with rich tracing, disciplined context and token control, native function binding, and provider-agnostic execution for long-running, headless agents.
-
Updated
May 14, 2026 - Kotlin
Long-horizon agent execution harness — reliable autonomous runs for Claude Code, Codex, OpenHands, and custom agents. Goal graphs, spin detection, HITL gates, fork/merge, 8 strategies, 6 validators.
-
Updated
May 1, 2026 - Python
bstack P12 — Persistent Loop Discipline. Cross-context restart loop with state in the filesystem. Closes the long-horizon context-rot failure mode. Composes with bstack P5/P6/P7/P10/P11.
-
Updated
May 6, 2026 - Python
Official repository for the paper: Can AI Agents Synthesize Scientific Conclusions?
benchmark ai-agents long-form long-horizon agentic-workflow scientific-conclusion-synthesis clean-room-evaluation
-
Updated
May 9, 2026
Improve this page
Add a description, image, and links to the long-horizon topic page so that developers can more easily learn about it.
Add this topic to your repo
To associate your repository with the long-horizon topic, visit your repo's landing page and select "manage topics."