LoreonLabsPlatform

Overview

Intelligence

Markets
Builders
Research
Narratives
Ecosystems
Launchpads

Discover

Search
Sources

Python

Multi-hop-Reasoning-VLM-Agent

wuxiyang1996/Multi-hop-Reasoning-VLM-Agent

PythonEmerging

Stars

—

Forks

—

Contributors

2

Last push

1mo ago

Recent commits

Latest commits.

feat(stage3): implement GRPO domain adaptation pipeline for 6 non-game benchmarks
4f44c53Dev1mo ago
feat(stage3): add 80/20 train/test splits for 6 non-game benchmarks
edc7527Dev1mo ago
feat(effects): close the contract_learn → intrinsic_bonus loop for seed mega-skills
497842bDev1mo ago
feat(coevo): bootstrap mega-skill contracts from teacher demos + v8/v9 launchers
7ec7ae8Dev1mo ago
feat: tetris XML state renderer, GRPO OOM resilience, training robustness

594979e

Dev

1mo ago

fix(coevo): Airstriker v7 — pure raw env reward, drop REASONING to stop CoT collapse

5a8c7daDev1mo ago

feat(stage2): Airstriker v6 — pivot vision perception to OpenRouter Gemini 2.5 Flash

2fd973aDev1mo ago

feat(stage2): add run_stage2_altered_beast_v6.sh launcher

344f692Dev1mo ago

Top contributors

Builders behind this project.