REVIEW 8 cited by
Bidirectional Decoding: Improving Action Chunking via Guided Test-Time Sampling
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Predicting and executing a sequence of actions without intermediate replanning, known as action chunking, is increasingly used in robot learning from human demonstrations. Yet, its effects on the learned policy remain inconsistent: some studies find it crucial for achieving strong results, while others observe decreased performance. In this paper, we first dissect how action chunking impacts the divergence between a learner and a demonstrator. We find that action chunking allows the learner to better capture the temporal dependencies in demonstrations but at the cost of reduced reactivity to unexpected states. To address this tradeoff, we propose Bidirectional Decoding (BID), a test-time inference algorithm that bridges action chunking with closed-loop adaptation. At each timestep, BID samples multiple candidate predictions and searches for the optimal one based on two criteria: (i) backward coherence, which favors samples that align with previous decisions; (ii) forward contrast, which seeks samples of high likelihood for future plans. By coupling decisions within and across action chunks, BID promotes both long-term consistency and short-term reactivity. Experimental results show that our method boosts the performance of two state-of-the-art generative policies across seven simulation benchmarks and two real-world tasks. Code and videos are available at https://bid-robot.github.io.
Forward citations
Cited by 8 Pith papers
-
SAIL: Faster-than-Demonstration Execution of Imitation Learning Policies
A full-stack speed-adaptation system lets imitation-learned robot policies execute up to 3-4x faster than human demonstrations while preserving task success rates.
-
ChainVLA: Chaining Vision-Language-Action Queries through a Unified Execution State for Long-Horizon Manipulation
ChainVLA chains vision-language-action queries by carrying both retrieved task progress and the previous prediction's unexecuted suffix into each new action horizon, reaching 62.8% on RMBench and 98.8% on LIBERO.
-
CheckVLA: Execution-Time Verification with Action-Conditioned World Model for Long-Horizon Mobile Manipulation
Action-conditioned world-model verification with conformal first-intervention control and latency-aware suffix repair raises RoboCasa365 success 8.5 points over invocation-matched periodic replanning.
-
TIDAL: Temporally Interleaved Diffusion and Action Loop for High-Frequency VLA Control
TIDAL raises VLA control feedback from ~2.4 Hz to ~9 Hz by caching semantic intent and interleaving one-step flow generation with execution, doubling dynamic interception success in simulation.
-
Temporal Action Selection for Action Chunking
TAS caches action predictions made at different past times and uses a small learned selector to choose which one to execute, improving success rates for action-chunking robot policies.
-
LLM-as-a-Verifier: A General-Purpose Verification Framework
Expecting over scoring-token logits yields continuous, scalable verification that improves agent trajectory selection and dense RL rewards across coding, robotics, and medical benchmarks.
-
Going with the Flow: Koopman Behavioral Models as Pseudo Planners for Visuo-Motor Dexterity
A single learned linear Koopman model over coupled visual and proprioceptive states generates full-horizon dexterous manipulation plans and triggers replanning when its own visual predictions diverge from reality.
-
Think Twice, Act Once: Token-Aware Compression and Action Reuse for Efficient Inference in Vision-Language-Action Models
FlashVLA, a training-free plug-in, reuses stable actions and prunes visual tokens to cut VLA model inference FLOPs by 55.7% and latency by 36% with only a 0.7% success-rate drop on LIBERO.
Discussion (0). Continue with ORCID to comment.