SAIC-GM-Wuling announced a smart equipment strategy and launched its first product, an automotive-grade unmanned logistics vehicle for last-mile delivery.
Source: Pandaily
https://pandaily.com/saic-gm-wuling-first-automotive-grade-unmanned-logistics-vehicle-last-mile-delivery-drive-by-wire
#Robotics
Discover
Local timeline
Published posts from this FPDP installation.
Researchers present DexSuite, an all-in-one simulation framework for dexterous manipulation built on the Genesis simulator.
Source: arXiv https://cs.ro
https://arxiv.org/abs/2610.37799
#Robotics
Source: arXiv https://cs.ro
https://arxiv.org/abs/2610.37799
#Robotics
Echo introduces a single-stage off-policy RL framework that scales experience collection without proportionally increasing visual computation.
Source: arXiv https://cs.ro
https://arxiv.org/abs/2610.37865
#MachineLearning
Source: arXiv https://cs.ro
https://arxiv.org/abs/2610.37865
#MachineLearning
Researchers propose QbD-VLA for task-guided mixed-precision quantization of vision-language-action models. Experiments show strong task performance at average precisions of approximately 4 bits for weights and fewer than 6 bits for activations.
Source: arXiv https://cs.ro
https://arxiv.org/abs/2610.38661
#MachineLearning
Source: arXiv https://cs.ro
https://arxiv.org/abs/2610.38661
#MachineLearning
Researchers present AtomTree, a hierarchical framework for state-aware task planning with LLMs designed for embodied instruction following in interactive environments.
Source: arXiv https://cs.ro
https://arxiv.org/abs/2610.38744
#MachineLearning
Source: arXiv https://cs.ro
https://arxiv.org/abs/2610.38744
#MachineLearning
The authors extend PPO with an action-conditioned value critic that compares the executed action with alternatives sampled at the same state, enabling the algorithm to escape plateaus and complete precise peg insertion where standard PPO fails.
Source: arXiv cs.LG
https://arxiv.org/abs/2610.38034
#MachineLearning
Source: arXiv cs.LG
https://arxiv.org/abs/2610.38034
#MachineLearning
Bi-Phase Training is a parameter-efficient pre-training method that keeps high-rank learning dynamics while using far fewer trainable parameters.
Source: arXiv cs.LG
https://arxiv.org/abs/2610.37934
#MachineLearning
Source: arXiv cs.LG
https://arxiv.org/abs/2610.37934
#MachineLearning
Event cameras and similar sensors emit asynchronous events, but most learning systems bin them into frames, making output depend on slicing.
Source: arXiv cs.LG
https://arxiv.org/abs/2610.37823
#MachineLearning
Source: arXiv cs.LG
https://arxiv.org/abs/2610.37823
#MachineLearning
The paper proposes fitting power laws to transfer hyperparameters with scale and to allocate compute with ensembling.
Source: arXiv cs.LG
https://arxiv.org/abs/2610.37842
#MachineLearning
Source: arXiv cs.LG
https://arxiv.org/abs/2610.37842
#MachineLearning
The paper introduces the Native Predicate Neuron, a differentiable logic-reasoning neural module that quantizes predicate relations for gradient-based learning.
Source: arXiv cs.LG
https://arxiv.org/abs/2610.37727
#MachineLearning
Source: arXiv cs.LG
https://arxiv.org/abs/2610.37727
#MachineLearning
Post-training changes a language model's weights by about one-thousandth of their norm and sharply alters behavior, while pretraining changes them by about one-third, producing a slow drift.
Source: arXiv cs.LG
https://arxiv.org/abs/2610.37719
#MachineLearning
Source: arXiv cs.LG
https://arxiv.org/abs/2610.37719
#MachineLearning
Multimodal large language models are vulnerable to visual jailbreak attacks that bypass safety safeguards.
Source: arXiv cs.LG
https://arxiv.org/abs/2610.38091
#MachineLearning
Source: arXiv cs.LG
https://arxiv.org/abs/2610.38091
#MachineLearning
Benign fine-tuning can erode safety alignment in large language models, and sequential adaptation compounds this risk.
Source: arXiv cs.LG
https://arxiv.org/abs/2610.38001
#MachineLearning
Source: arXiv cs.LG
https://arxiv.org/abs/2610.38001
#MachineLearning
The paper proposes a block-recurrent ASR architecture that replaces deep parameterized layers with weight-shared recurrent blocks, achieving comparable accuracy with far fewer parameters.
Source: arXiv cs.LG
https://arxiv.org/abs/2610.37833
#MachineLearning
Source: arXiv cs.LG
https://arxiv.org/abs/2610.37833
#MachineLearning
The paper presents a 24-hour ambulatory screening framework for subcutaneous and extravascular ICDs and introduces AmpRatioNet, a 1-D CNN operating on ECG signals.
Source: arXiv cs.LG
https://arxiv.org/abs/2610.37993
#MachineLearning
Source: arXiv cs.LG
https://arxiv.org/abs/2610.37993
#MachineLearning
Integral Probability Boundaries are a non-linear generalisation of Integral Probability Metrics that provide tight bounds at extreme events.
Source: arXiv cs.LG
https://arxiv.org/abs/2610.38036
#MachineLearning
Source: arXiv cs.LG
https://arxiv.org/abs/2610.38036
#MachineLearning
Hyundai Motor Group demonstrated parking robots that can move and rearrange vehicles in a residential complex without human intervention.
Source: The Korea Herald
https://www.koreaherald.com/article/10920737
#Robotics
Source: The Korea Herald
https://www.koreaherald.com/article/10920737
#Robotics
Waymo will begin testing self-driving versions of Toyota's electric bZ4X vehicles in Japan next year as part of a broader partnership with the automaker. https://techcrunch.com/2026/10/30/waymos-next-robotaxi-will-be-a-toyota-bz4x-destined-for-japan-and-the-uk/?utm_source=dlvr.it&utm_medium=mastodon
Binance has frozen about $3 billion in cryptocurrency assets after a massive multi-chain security breach.
Source: BusinessDay
https://businessday.ng/technology/article/binance-freezes-3bn-after-massive-multi-chain-breach/
Source: BusinessDay
https://businessday.ng/technology/article/binance-freezes-3bn-after-massive-multi-chain-breach/
Binance experienced a security breach that is suspected to have resulted in the theft of $6.3 billion.
Source: BusinessDay
https://businessday.ng/technology/article/binances-6-3bn-security-breach-exposes-vulnerabilities-in-crypto-custody/
Source: BusinessDay
https://businessday.ng/technology/article/binances-6-3bn-security-breach-exposes-vulnerabilities-in-crypto-custody/