arXiv:2606. 06660v1 Announce Type: new Abstract: Long-horizon robot manipulation tends to fail gradually: one bad step degrades the state, and the policy spirals into a basin from which it cannot recover.
arXiv:2610.32941v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models perform well on robotic manipulation, yet remain brittle once their own actions carry them into states their trai...
arXiv:2608. 03222v1 Announce Type: cross Abstract: Software engineering (SWE) agents resolve repository-level issues through long trajectories that grow increasingly expensive as context accumulates.
arXiv:2608. 08239v1 Announce Type: new Abstract: LLM routers promise efficiency by matching each request to the cheapest adequate model, and are increasingly applied per step inside multi-step agents.
The paper introduces RESCUE, a counterfactual framework that evaluates whether runtime monitors for robot policies generate alarms that lead to useful interventions. By restoring simulator and policy states at checkpoints, RESCUE compares alternative actions to identify the latest checkpoint that satisfies a recovery criterion and measures whether acting on alarms reduces expected task‑completion cost. Experiments on OpenVLA‑OFT and π0.5 over LIBERO‑Spatial and LIBERO‑Plus show that recoverability depends on the state–intervention pair, and that even an ideal detector can only tolerate a 2.0–3.3% false‑alarm rate before unnecessary resets outweigh early failure stops.
Software engineering (SWE) agents resolve repository-level issues through long trajectories that grow increasingly expensive as context accumulates. Failed runs tend to be longer and exhibit redundant exploration or looping, suggesting that some failures may be detectable before completion.
The paper investigates Observation‑Space Shift (OSS), a failure mode in long‑horizon robotic manipulation where downstream skills start from states altered by preceding skills. By diagnosing that displaced scene elements—not robot joints or target objects—cause most OSS, the authors develop a detect‑restore‑resume system that monitors task progress, restores the scene, and fine‑tunes the skill to resume. On the BOSS‑44 benchmark, this approach raises full‑chain success from 7.6 % to 26.5 %, outperforming several baselines and demonstrating the value of scene restoration before policy resumption.
The paper introduces instruction duplication, a simple inference‑time control that repeats the procedural instruction without retraining or decoding changes. Across seven instruction‑tuned models and 16,800 scheduled generations, duplicating the instruction improves deterministic All‑8 diagnostic‑response success from 90.22% to 93.17% and reduces failures by 30.2%. In downstream Answer Engineering scenarios, duplication further boosts success rates, demonstrating its practical impact on systems that rely on the generated trajectory.
The paper introduces CARE, a method for certifying accelerated inference in vision‑language‑action models. CARE uses paired rollouts on a calibration set to guarantee that the risk of acceleration‑induced failure stays below a user‑specified budget, selecting the fastest certified accelerator or falling back to the reference policy. Experiments on LIBERO suites show that CARE achieves 9.0–10.8× speedups while preserving at least 85.8% of reference‑solved episodes with 95% confidence, and it outperforms non‑guaranteed selectors in both budget adherence and rollout efficiency.
The paper investigates a specific failure mode in vision‑language‑action models, termed instruction‑action binding, where models respond to language and vision separately but fail to combine them to select the correct action under counterfactual changes. Through behavioral analyses of fine‑tuned policies, the authors show that failed rollouts often preserve source behavior or switch to other demonstrated tasks, indicating that language is not ignored but mis‑bound. They propose Equivariant Counterfactual Training (ECT), which supplies counterfactual demonstrations and a paired loss to enforce correct action selection, achieving significant performance gains across simulated and real‑world benchmarks.
arXiv:2608. 02464v1 Announce Type: cross Abstract: LLM agents fail mid-episode -- they loop, cascade tool errors, drift off goal, fabricate results, or silently absorb corrupted content -- and the standard remedy, judging every step with a second LLM, costs more than the agent itself.
The paper investigates the safety of Code World Models, where a language model generates executable world models that a planner uses. It shows that accepting a model based on sampled transitions only guarantees sample consistency, not full safety, because the probability of missing critical events decays as (1‑r)^N. Experiments on hybrid instruments reveal that omitted mode‑boundaries can severely limit planner performance, and that even sophisticated LLMs (GPT‑5.x) struggle to repair such omissions in higher‑dimensional settings.
Instruction duplication is a simple, inference‑time control that repeats the procedural instruction in a language‑model output without retraining or decoding changes. In experiments across seven instruction‑tuned models on 300 medical multiple‑choice questions, duplicating the instruction increased the proportion of deterministic All‑8 diagnostic‑responses from 90.22 % to 93.17 % and reduced failures by 30.2 %. The technique also improved pre‑provisional TF‑IDF recall and, in downstream Answer Engineering scenarios, significantly raised success rates for specific endpoints.