DeepJEPA is a weight‑tied joint‑embedding predictive world model that treats transition depth as an inner test‑time scaling axis, learning when additional recurrent updates are worthwhile for each candidate and rollout step. Unlike traditional planners that uniformly deepen every transition, DeepJEPA concentrates extra computation on decision‑critical events such as contact onset and sustained object interaction, achieving comparable or better performance with only 1.00–1.26 updates per transition across five visual‑control settings. The approach demonstrates that improved planning does not require uniformly better object‑state decodability, but rather targeted internal computation where it can alter the planner’s elite set and action selection.
By Zijian Jin, Yunbei Zhang, Yuanzhe Liu, Ming Liu, Baian Chen, Weirui Ye, Shilong Liu, Marco Pavone
The paper introduces token‑wise incremental trajectories, a method that tracks how contextual word embeddings evolve as each word is added to a sentence. Using garden‑path sentences, the authors show that these trajectories capture known processing disruptions and can distinguish ambiguous sentences from disambiguated controls. They propose metrics for measuring representational displacement and find that ambiguity signals are present not only in sentence‑level CLS vectors but also in individual token embeddings, suggesting distributed meaning construction across scales.
By Grayson Wycliffe Storer, Julia Witte Zimmerman
The study investigates whether vision–language models (VLMs) can infer human engagement from gameplay videos using the GameVibe Few‑Shot dataset across nine first‑person shooter games. Three VLMs were tested under six prompting strategies—including zero‑shot, theory‑guided prompts based on Flow, GameFlow, Self‑Determination Theory, and MDA, and retrieval‑augmented prompting—evaluating both pointwise engagement prediction and pairwise prediction of engagement change. Results show that zero‑shot predictions are weak and often do not beat simple majority‑class baselines; retrieval‑augmented prompting improves pointwise prediction in some cases, while pairwise prediction remains difficult, and theory‑guided prompts do not reliably help and may reinforce superficial shortcuts.
"whyItMatters":"The findings highlight a perception–understanding gap in current VLMs, indicating that while they can recognize visible gameplay cues, they still struggle to robustly infer human engagement across games."
By Ziyi Wang, Qizan Guo, Rishitosh Singh, Xiyang Hu
The paper presents C2C (From Codebase to Culprit), a bug‑localization framework that narrows the debugging search space through semantic retrieval and hierarchical reinforcement learning. It first retrieves likely buggy files using contrastively fine‑tuned CodeBERT embeddings and then progressively localizes bugs from files to functions to lines of code. Experiments on Java and Python datasets show that this multi‑resolution approach improves both retrieval precision and localization accuracy, with ablation studies underscoring the importance of hierarchical decomposition and reward shaping.
By Ankur Garg, Corey Yang-Smith, Rishav Rishav, Ahmad Abdellatif, Samira Ebrahimi Kahou
GPEC is a lightweight pre‑LLM Gaussian Process Embedding Correction that refines visual representations for cardiac ultrasound caption generation. By inserting a residual correction layer between the visual projection and the language model, it aligns projected embeddings with annotation‑guided targets derived from structured video annotations. Experiments show that GPEC improves caption similarity and content alignment while adding less than 0.05 s of inference overhead, all without fine‑tuning the pretrained multimodal backbone.
By Arefeh Rezaei
The paper introduces Align Then Reason (ATR), a multilingual lip‑sync judge that aligns frame‑level lip representations with phonetic units of a candidate text line and then uses a language model to evaluate both content and timing. ATR achieves significant improvements over existing baselines on a seven‑language benchmark, with mean AUC gains of up to 59.4% for 2B reasoners and similar gains across other LLM families. The method also transfers well to unseen languages and outperforms lip‑reading baselines on real dubbing tasks such as dub‑line reranking and script‑to‑clip assignment.
By Rui Liu, Bhavin Jawade, Haoqi Li, Shivam Mehta, Karan Saxena, Yinghong Lan, Cameron R. Wolfe
Watch Your Speech (WYS) is a video‑to‑speech synthesis framework that uses textual conditioning to resolve the one‑to‑many mapping problem inherent in silent talking‑face videos. It fuses textual context with video sequences via an attention‑based embedding module and employs a conditional flow matching objective to produce high‑fidelity, phonemically accurate speech. Experiments on LRS2 and LRS3 show that WYS sets new state‑of‑the‑art results in audio‑visual synchronization while maintaining competitive word‑error rates, and subjective tests confirm near‑human naturalness.
By Gunwoo Lee, Yoori Oh, Yoseob Han
SlotVLA proposes a slot-attention framework that models both objects and their relations for robotic manipulation, using a relation‑centric decoder and an LLM‑driven action translator. The authors introduce LIBERO+, a fine‑grained benchmark with object‑centric annotations, box and mask labels, and instance‑level temporal tracking to evaluate object‑relation reasoning. Experiments on LIBERO+ show that object‑centric and object‑relation slot representations reduce visual token counts while maintaining competitive generalization, offering a compact and interpretable foundation for multitask visuomotor control.
By Taisei Hanyu, Nhat Chung, Huy Le, Toan Nguyen, Yuki Ikebe, Anthony Gunderman, Duy Nguyen Ho Minh, Khoa Vo, Tung Kieu, Kashu Yamazaki, Chase Rainwater, Anh Nguyen, Ngan Le
SpikeMoE introduces a spike-based k‑WTA router that uses lateral inhibition and refractory periods to select the top‑K experts based on discrete spike counts, inspired by hippocampal CA1 competition. The framework combines spiking neural network dynamics with mixture‑of‑experts conditional computation and adds a two‑stage missing‑modality module for robust multimodal processing. Experiments on vision, language, and multimodal tasks show that SpikeMoE matches or surpasses ANN baselines while offering energy‑efficient performance.
By Xiaoli Liu, Yujie Liang, Jialin Li, Malu Zhang
MOMAT (Mixture of Multiple Atlases) is a hardware‑enhanced safety framework designed to defend quantized large language models (qLLMs) on edge devices against jailbreak attacks. It uses a collection of semantic atlases—each containing harmful or benign sample clusters and policy templates—to perform domain‑localized Retrieval‑Augmented Generation. A lightweight Mixture of Experts detector evaluates top‑k similarity features retrieved by a Compute‑in‑Memory (CiM) accelerated engine, achieving a 4.69 × 10⁶‑fold speedup and a 2.5 × 10⁵‑fold energy reduction compared to DRAM‑based baselines while matching state‑of‑the‑art defense performance.
By Boyang Li, Bingyu Shen, Weihao Hong, Zhiyuan Jiang, Xinlei Guan, Yan Ma, Miles Q. Li, Yi Sheng, Ruiyang Qin
The paper introduces LOOM, a method for scaling looped mixture‑of‑experts (MoE) Transformers beyond the typical two‑loop limit. LOOM addresses two key obstacles: it stabilizes deep recurrence by bounding residual variance and re‑injecting the input embedding, and it prevents expert selection collapse by using per‑loop routers and a looping residual to maintain computational diversity. Experiments on 100 M–1.7 B parameter models show stable scaling to 9–12 loops, with significant perplexity reductions and zero‑shot accuracy gains under near‑iso‑FLOP conditions.
By Di He, Pengxiang Li, Da Chang, Qingyan Meng, Lu Yin, Shiwei Liu
The paper introduces TRACE, a method that removes duplicated text—known as note bloat—from clinical notes by leveraging EHR metadata and frequency-based de‑duplication. Across 5.3 million notes from diverse patient cohorts, TRACE eliminated 47.3 % of chart text while preserving information extraction and prediction performance, with only 0.3–6.6 % of removed content being author‑generated. The authors project that applying TRACE could yield net savings of $1.00 M to $13.58 M over three years at a large academic center, depending on model pricing schemes.
By Jordan L. Cahoon, Chloe Stanwyck, Asad Aali, Rachel Madding, Sulaiman S. Somani, Emma Sun, Yixing Jiang, Renumathy Dhanasekaran, Emily Alsentzer
The paper introduces T-ARC, a clustering algorithm that integrates topological information into the K‑means objective by coupling a data‑fidelity term with a graph‑cut penalty. The latent graph is modeled as a random realization from a Stochastic Block Model, whose parameter is optimized via Distributionally Robust Optimization, using a persistence‑based similarity matrix derived from zero‑dimensional persistent homology. Experiments on synthetic non‑convex data and Fashion‑MNIST subsets demonstrate that T‑ARC recovers latent topological structures and outperforms K‑means on curved and interleaved clusters while remaining competitive and more stable on real data.
By Serena Grazia De Benedictis, Andersen Ang, Nicoletta Del Buono, Flavia Esposito, Laura Selicato
STAGE is a training‑free, closed‑form framework for concept erasure in native text‑to‑3D generators. It treats erasure as a stage‑aware editing problem, applying low‑dimensional affine corrections separately to the structural and appearance stages of the pipeline. Experiments on the TRELLIS generator show that STAGE outperforms adapted baselines, achieving a composite score of 66.7 versus 53.2 across 15 shape, material, and object concepts.
By Karol Dziekan, Przemys{\l}aw Spurek, Dawid Malarz
The paper introduces a pipeline that automatically creates ontology‑grounded multiple‑choice question benchmarks for evaluating large language models (LLMs) on logical reasoning tasks in scientific AI. By using OWL 2 ontologies, correct answers are guaranteed by design and distractors are generated and formally verified as incorrect through an OWL reasoner. Experiments on three ontologies—Pizza, PMDco, and DOID—yielded 112, 2,491, and 15,216 MCQs, respectively, with high natural‑language quality and challenging zero‑shot performance for six LLMs.
By Nishtha N. Vaidya, Stephan Grimm, Thomas Hubauer, Thomas A. Runkler
The paper investigates how small action errors evolve when using action chunking in behavioural cloning. By injecting errors at each state and observing their growth under open‑loop (no replanning) and closed‑loop (replanning) regimes, the authors classify states as contracting, expanding, or unresolved. Across twelve manipulation tasks, they find that stable states are rare, error amplification is common, and that short‑horizon fitting can overestimate long‑horizon propagation. Predictors trained on camera and proprioceptive data can recover open‑loop stability but only partially capture closed‑loop dynamics, indicating that standard imitation learning does not reliably produce policies that contract errors when perturbed.
By Aryan Goyal
AVSD-Scenes is a new dataset of 12,291 audio‑visual scene descriptions for urban environments, built from the TAU Urban Audio‑Visual Scenes dataset. The descriptions are generated by first creating modality‑specific text with Qwen2‑Audio‑7B and Qwen2.5‑VL‑7B, then merging them with large language models (Qwen3‑14B, Mistral‑Small‑3.2‑24B‑Instruct‑2506, Gemma‑3‑27B‑it) to produce multimodal narratives that combine auditory and visual cues. Benchmarks show that these multimodal descriptions improve semantic alignment, cross‑modal retrieval, and scene classification accuracy (up to 95.4%) while remaining discriminative even without explicit scene labels.
By Dhanunjaya Varma Devalraju, Arshdeep Singh, Mark D. Plumbley
The paper surveys recent advances in Retrieval Augmented Generation (RAG), a technique that integrates external retrieval into language model generation to reduce hallucinations and keep knowledge current. It introduces a four‑axis taxonomy—efficiency, defense, interactivity, and reasoning—to organize contemporary RAG research, covering retrieval methods, fusion strategies, embedding optimizations, and reinforcement learning policies. The survey also reviews evaluation practices, domain‑specific applications, and architectural variants, while highlighting ongoing challenges such as retrieval quality, reliability, domain adaptation, scalability, and explainability.
By Meghana Sunil, Shravya V, Shravan Venkatraman, Joe Dhanith PR
Omni-Embed-Mini is a 0.9B‑parameter model that embeds text, speech, audio, images, video, and visually‑rich documents into a single shared cosine space without updating any text‑side parameters. It uses a dense cascaded caption as a teacher signal, allowing the teacher and student to share identical backbone weights and requiring only lightweight projectors and phased LoRA adapters for alignment. The model achieves strong text retrieval performance (49.57 nDCG@10 on MTEB‑v2 BEIR‑8) while extending to five additional modalities and is significantly smaller than other open omni‑modal embedders.
By Mohammed Irfan Kurpath, Jaseel Muhammad Kaithakkodan, Sahal Shaji Mullappilly, Ivan Laptev, Hisham Cholakkal
MemFit is a long‑term memory system designed for conversational agents that stores each dialogue turn verbatim in an append‑only store, enabling near‑instantaneous, LLM‑free insertion. It indexes turns using segment summaries and employs an LLM‑free, multi‑path retrieval strategy that blends lexical and semantic signals with cross‑encoder reranking over caption‑augmented episodes. Experiments on LoCoMo, MemGallery, and LongMemEval‑S demonstrate state‑of‑the‑art performance while drastically reducing memory construction time and cost.
By Mitchell Piehl, Muchao Ye