Event-grounded SAE analysis in VLA policies produces stronger causal effects on robot behavior than standard methods by anchoring features to clustered end-effector keyframes across simulations and real-robot tests.
When vision overrides language: Evaluating and mitigating counterfactual failures in vlas
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.RO 4years
2026 4roles
background 1polarities
background 1representative citing papers
LA4VLA creates a 33K language-action dataset from existing demos and shows that pretraining on language-action pairs before or alongside vision-language-action training boosts success rates in sim and real robot tasks.
RoboSemanticBench reveals that representative VLA models grasp blocks successfully but select the semantically correct answer at near-random rates, indicating a gap between backbone semantics and action prediction.
VLAs-as-Tools pairs a VLM planner with specialized VLA executors via a new interface and Tool-Aligned Post-Training to raise long-horizon robot success rates on LIBERO-Long and RoboTwin benchmarks.
citing papers explorer
-
LA4VLA: Learning to Act without Seeing via Language-Action Pretraining
LA4VLA creates a 33K language-action dataset from existing demos and shows that pretraining on language-action pairs before or alongside vision-language-action training boosts success rates in sim and real robot tasks.
-
RoboSemanticBench: Diagnosing Semantic Grounding in Action Prediction for VLA Models
RoboSemanticBench reveals that representative VLA models grasp blocks successfully but select the semantically correct answer at near-random rates, indicating a gap between backbone semantics and action prediction.
-
Towards Long-horizon Embodied Agents with Tool-Aligned Vision-Language-Action Models
VLAs-as-Tools pairs a VLM planner with specialized VLA executors via a new interface and Tool-Aligned Post-Training to raise long-horizon robot success rates on LIBERO-Long and RoboTwin benchmarks.