When Model Merging Rivals Joint Multi-Task Reinforcement Learning: A Task-Vector Geometry Analysis
Model merging is promoted as a substitute for joint multi-task training, yet in the reinforcement-learning setting this substitution is essentially never tested against the baseline it claims to replace: methods merge independently released agents precisely because a joint model is unavailable. We build the missing comparison. Training difficulty-1 and difficulty-2 Qwen3-8B specialists on the AppWorld agent benchmark with LOOP, we merge them (TIES, RAM+) and pit the result against a jointly trained model on the same data. On task-goal completion, merging matches joint RL -- and every merge var
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- PossiblePossibly related (embedding) · 48%I RL-trained Qwen3.6-35B-A3B to RL-train small task-specific Qwen models. Fully open source! 🤓 →
- PossiblePossibly related (embedding) · 46%RL without TD learning →
- FuzzySimilar title/name (fuzzy) · 59%aymericdamien/TopDeepLearning →
“Fuzzy title match (0.73): “When Model Merging Rivals Joint Multi-Task Reinforcement Lea” ≈ “aymericdamien/TopDeepLearning””
- LinkedLinked via arxiv author · 85%S. Aaron McClendon →
“When Model Merging Rivals Joint Multi-Task Reinforcement Learning: A Task-Vector Geometry Analysis”