The paper proposes a three‑stage training pipeline that begins with procedural pretraining on abstract, procedurally generated data, followed by molecular pretraining on SMILES, and finally downstream fine‑tuning for molecular property prediction. Experiments show that procedural pretraining improves downstream performance—e.g., a 4.8% error reduction on Lipophilicity—especially when labeled data are scarce, and that the benefit peaks at an intermediate procedural training budget. Analysis indicates that transferable knowledge resides mainly in attention layers, while feed‑forward layers may over‑specialize.
By Moritz Friedemann, Zachary Shinnick, Philip Torr, Bruno Andreis
The paper introduces CUES, a lightweight heuristic for selecting encoder combinations in large audio‑language models by estimating complementarity through Pearson correlations of single‑encoder performance profiles. Using a frozen SmolLM2‑135M backbone, CUES consistently identifies optimal encoder sets for each track on the XARES‑LLM benchmark without requiring fusion training or test data. On broad audio tasks, CUES selects a diverse trio of encoders, improving performance by 4.3% over Whisper‑medium, while on text generation it opts for a focused speech‑only pair, outperforming mHuBERT‑147 by 6.3%. The results illustrate how correlation signals guide a diversity–interference trade‑off across different task families.
By Pei-Jun Liao, Hung-Shin Lee, Wenze Ren, Kuo-Hsuan Hung, Hung-yi Lee, Hsin-Min Wang
arXiv:2609.17548v1 Announce Type: new
Abstract: Myovox, from myo (muscle) and vox (voice), decodes open-vocabulary English text from 31-channel surface electromyography (sEMG) recorded from the muscl...
By Varshith Madishetty
arXiv:2609.17848v1 Announce Type: new
Abstract: Limited controlled evidence exists on how training data, adaptation method, and model scale jointly affect tool-calling performance in language-model a...
By Md Tahmid Rahman Laskar, Xue-Yong Fu, Shashi Bhushan TN
The study investigates whether adding specialized modules—cross‑depth Attention Residuals (AttnRes) and soft‑routed expert banks—to a fully fine‑tuned EEG foundation model improves performance on various EEG decoding tasks. Across four datasets (FACED, ISRUC, SEED‑V, PhysioNet‑MI), AttnRes alone boosts mean balanced accuracy on three datasets, while the full model with experts only benefits two. However, these gains come with significant computational overhead, and the effects are dataset‑dependent rather than consistently positive.
By Mingyang Jiang, Yamin Li, Daniel Moyer, Fan Ma, Hua Xu, Catie Chang
arXiv:2609.19142v1 Announce Type: new
Abstract: World models endow perceptual systems with the ability to predict how scenes evolve under interaction. They are most beneficial when trained on diverse...
By Bardienus P. Duisterhof, Kaifeng Zhang, Adam Hung, Bowen Wen, Stan Birchfield, Yunzhu Li, Deva Ramanan, Jeffrey Ichnowski
arXiv:2609.17539v1 Announce Type: new
Abstract: We present MudawanSn, a gold-standard resource of 1,271 sentence-aligned pairs manually translated from Wolof into Modern Standard Arabic (MSA). The so...
By Mouhamed Mbaye, Thierno Diop
This paper introduces an energy-aware approach to robotic manipulation by defining a joint-space mechanical-work proxy based on joint torque and angular displacement. A differentiable energy predictor is trained to estimate this work from robot states and actions, enabling it to serve as a regularizer that fine‑tunes a pretrained manipulation policy. Applied to RVT‑2 on RLBench, the method reduces average mechanical work from 208.8 J to 204.4 J (a 2.1 % drop) while slightly improving task success from 86.2 % to 86.9 % across 12 manipulation tasks.
By Toshiki Otani, Hiromu Taketsugu, Norimichi Ukita
FRAUDSkill is a structured frozen‑weight adaptation framework for audio anti‑fraud detection that keeps the underlying audio‑language model unchanged while optimizing external skill programs, route‑specific policies, and decision rules. It combines structured output control with validation‑guided multi‑path inference to produce protocol‑compliant predictions. On the TeleAntiFraud benchmark, FRAUDSkill achieves a 73.50% Macro‑F1 score, outperforming the shared frozen‑model baseline by 31.96% and reducing invalid outputs to 1.94%.
By Chengxian Hu, Zhiming Ma, Mingjun Pan, Yifan Wang, Shun Zhang, Qifan Wang, Zhilei Zhao, Yijin Zhou, Yuxi Zhao, Huiyuan Liu, Peidong Wang, Peng Chen
The report introduces Nemotron-SEA-LION-v4.8, a family of Southeast Asian language models built on NVIDIA Nemotron 3, featuring 30B-A3B and 120B-A12B variants with both base and post‑trained checkpoints. The models are fine‑tuned on Southeast Asian, reasoning, code, and multilingual parallel datasets, then further refined with supervised fine‑tuning and online on‑policy distillation. On the SEA‑HELM benchmark, the 30B-A3B model raises the overall SEA score from 46.06 to 51.57, while the 120B-A12B model jumps from 49.30 to 63.44, with the largest improvements seen in instruction following, natural language reasoning, and understanding across seven Southeast Asian languages.
By Ahmed Mohammad Dabeer (David Wang Dawei), Ahn Jeongmi (David Wang Dawei), Anocha Sutaveephamochanon (David Wang Dawei), Antonyrex Sajeban (David Wang Dawei), Aulia Adila (David Wang Dawei), Chan Hok Teng (David Wang Dawei), Adwin (David Wang Dawei), Cheng Zi Yi (David Wang Dawei), Nicholas Zhuang Ziyi (David Wang Dawei), Choa Hsueh Mei Esther (David Wang Dawei), David Ong Tat-Wee (David Wang Dawei), Evelyn Tan Chor Phin (Li Chunren), Heng Cheng Peng (Li Chunren), Jonathan (Li Chunren), Lee Chwan Ren (Li Chunren), Leong Wai Yi (Huang Wenzong, Raymond), Leong Wei Qi (Huang Wenzong, Raymond), Leslie Teo Eng Sipp (Huang Wenzong, Raymond), Liew Rachel (Huang Wenzong, Raymond), Limkonchotiwat Peerat (Huang Wenzong, Raymond), Montalan Jann Railey Estrada (Huang Wenzong, Raymond), Muhammad Ridzuan Bin Mokhtar (Huang Wenzong, Raymond), Nagarajan Karthik (Huang Wenzong, Raymond), Ng Boon Cheong (Huang Wenzong, Raymond), Raymond (Huang Wenzong, Raymond), Ngui Jian Gang (Chen Xiaowei), Nguyen Thanh Ngan (Chen Xiaowei), Tasawong Panuthep (Chen Xiaowei), Pereira Mark Gregory (Chen Xiaowei), Phang Shi Wei Benjamin (Chen Xiaowei), Poon Yip Hung (Chen Xiaowei), Joseph (Chen Xiaowei), Rengarajan Hamsawardhini (Chen Xiaowei), Siow Wei Kang Bryan (Chen Xiaowei), Tai Ngee Chia (Chen Xiaowei), Tan Choon Meng (Chen Xiaowei), Tan Le Min (Chen Xiaowei), Sheryl (Chen Xiaowei), Tan Siao Wei (Chen Xiaowei), Tan Yi Xian, Tee Jun Yun, Teng Kok Wai, Tjhi William Chandra, Tuchinda Pume, Wu Donghang, Yong Xianbin, Yosephine, Zhang Zhou
TabICLv2 is a new state‑of‑the‑art tabular foundation model that outperforms existing methods on regression and classification tasks. It relies on a synthetic data generation engine for diverse pretraining, architectural innovations such as a scalable softmax attention, and optimized training protocols that replace AdamW with the Muon optimizer. On the TabArena and TALENT benchmarks, TabICLv2 surpasses the current best model, RealTabPFN‑2.5, without any tuning, while also being faster and capable of handling million‑scale datasets with limited GPU memory.
By Jingang Qu, David Holzm\"uller, Ga\"el Varoquaux, Marine Le Morvan
The paper investigates how to balance model size and fine‑tuning strategy for UAV audio classification. Using a dataset of 3,100 clips across 31 drone classes, it compares transformer and convolutional backbones under full fine‑tuning, classifier‑only fine‑tuning, and four parameter‑efficient fine‑tuning methods. Results show that selective batch‑norm tuning of EfficientNet‑B7 yields the best accuracy (97.65%) while updating less than 0.5% of parameters, and that lightweight CNNs generally outperform transformers in both accuracy and efficiency.
By Andrew P. Berg, Qian Zhang, Mia Y. Wang
GOLF is a method that won the SHOW3D Interaction Field Estimation Challenge at HANDS@ECCV 2026. It predicts a 3D vector from each of 21 hand joints to the nearest point on a manipulated object using synchronized egocentric stereo views. The approach fuses dense global context, locally sampled hand/object evidence, and Plücker‑ray geometry, and employs a DINOv3 ViT‑H+/16 backbone with LoRA and trainable LayerNorm, achieving top scores on the hidden test set.
By Minqiang Zou, Riqiang Jin, Zhi Lv, Dong Luo, Lianghai Tian, Zhenyu Zhao, Qi Xu, Tong Wu, Mochen Yu, Yao Tang
BadQubits is an LLM-based framework that statically analyzes OpenQASM 2.0 circuits before execution to detect structurally harmful patterns. The system evaluates four large language model architectures on 1,500 circuits, achieving 92.67% classification accuracy and 96.1% recall for harmful circuits with a fine‑tuned Qwen Coder 2.5 7B model. Comparative experiments show that LLMs retain sequential token structure, outperforming a bag‑of‑gates CNN, and that model decisions correlate with threat‑defining features such as SWAP density and measurement timing.
By Justin Woodring, Lamine Noureddine, Aisha Ali-Gombe
arXiv:2609.18971v1 Announce Type: new
Abstract: Surgical phase recognition maps each video frame to a clinically meaningful workflow phase, supporting context-aware assistance, documentation, and pos...
By Ye Tao, Claudia Scherl, Sara Monji-Azad
arXiv:2609.18680v1 Announce Type: new
Abstract: Contextual ASR can benefit from semantic cues or from target words explicitly provided in the context. We introduce HearInContext, a Mandarin--English...
By Yifan Gao, Yao Tian, Hongbin Suo
This paper investigates instrument classification using solo sheet music images rather than audio. It converts images into sequences of musical words via bootleg score representation and treats the task as text classification, training AWD‑LSTM, GPT‑2, and RoBERTa models on IMSLP data for eight instruments. Pretraining on unlabeled data and fine‑tuning improves RoBERTa’s accuracy from 34.5% to 42.9%, and two proposed data‑augmentation methods raise accuracy by an additional 15%.
By Kevin Ji, Daniel Yang, TJ Tsai
The paper introduces SARA, an industrial framework that scales articulated user rationales (AURs) for recommendation systems. It curates a high‑quality AUR dataset from 240 M users, trains a 7B‑parameter MLLM (SARA‑7B) to generate rationales for millions of authors, and integrates these generated rationales into a production ranking model (SARA‑Ranker). Offline and online experiments demonstrate that the system produces more specific, polarity‑consistent rationales and improves user engagement while reducing negative feedback.
By Haoke Xiao, Yueyang Liu, Yuhui Zhang, Xiang Chen, Yufei Liu, Jia Xu, Yalong Guan, Xiaolan Zhu, Xiaoyu Zhang, Shijun Wang, Shuang Yang, Zijie Meng, Zejian Zhang, Ruochen Yang, Xiangyu Wu, Tingting Gao, Han Li, Lantao Hu, Cheng Luo, Kun Gai
The paper investigates why large vision‑language models sometimes misclassify harmful memes, attributing failures to either missing internal evidence or poor routing of evidence to the output. Using sparse autoencoders, role‑conditioned probes, and causal interventions on Gemma‑3 and Qwen3.5, the authors show that sparse readouts consistently outperform native predictions across six harmful content benchmarks, revealing a readout gap that is largely due to routing rather than representation. The study also demonstrates that calibration‑only routing recovers most of the performance gap and that the issue persists across languages and is not solely driven by OCR signals.
By Girish A. Koushik, Diptesh Kanojia, Helen Treharne
STRADAViT is a self‑supervised continued‑pretraining framework that adapts Vision Transformer (ViT) backbones for radio‑astronomy image analysis. It curates mixed‑survey data, generates radio‑astronomy‑aware training views, and initializes encoders with ViT‑MAE, optionally adding register tokens. Evaluations on three morphology benchmarks (MiraBest, LoTSS DR2, and Radio Galaxy Zoo) show that a register‑based two‑stage checkpoint improves linear‑probe Macro‑F1 scores over the ViT‑MAE baseline and enhances fine‑tuning on MiraBest and RGZ DR1, though performance on LoTSS DR2 fine‑tuning declines; these differences are statistically significant.
By Andrea DeMarco, Ian Fenech Conti, Hayley Camilleri, Ardiana Bushi, Simone Riggi