Unverified50% confidenceFactExact time
The paper's experiments found that the supervision method that performs best on the source benchmark is not necessarily likewise optimal on the transfer benchmark, so evaluating the teaching utility of reasoning traces needs to consider cross-distribution generalization.
1
Sources
50%
Confidence
Long-term
Relevance
10/19/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedEvaluating multimodal embedding training strategies must return to unified benchmarks and reproducible conditions; improvements reported by different papers under their own settings are often difficult to compare horizontally, and only controlled experiments within a unified framework can expose the true contribution of a strategy.75% similarUnverifiedEvaluating within the training distribution easily overestimates model capability, while out-of-distribution benchmarks require the model to generalize to unseen scene and interaction combinations, better exposing the gap between 'looking right' and 'being physically right'.73% similarUnverifiedReinforcement learning in the post-training stage can improve answer accuracy on both in-distribution and out-of-distribution tasks, but does not consistently improve the validity of reasoning traces.72% similarUnverified参数量相似的情况下，训练数据质量分布的差异往往比架构差异更能决定特定场景的表现上限72% similarVerifiedBehavioral cloning suffers from the distribution shift problem: during training the policy only encounters the state distribution covered by expert trajectories, and during deployment the accumulation of small deviations causes the actual state distribution visited to gradually diverge from the training distribution.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1064848API
curl https://kongchang.com/api/v1/knowledge/claims/1064848MCP
get_claim(id=1064848)