Leaderboards, eval harnesses and ablations — the contested business of deciding which model is actually better.
arXiv:2605.28013v2 Announce Type: replace Abstract: Multimodal Large Language Models (MLLMs) exacerbate safety risks by introducing vulnerabilities across multiple modalities, such as language and vi...
arXiv:2609.36145v1 Announce Type: new Abstract: Tampered Text Detection (TTD) is essential for safeguarding document authenticity in security-critical workflows. Existing expert models are effective...
arXiv:2609.36189v1 Announce Type: new Abstract: Routine chest CT captures upper-abdominal structures that may contain clinically relevant incidental abnormalities. Detecting these findings requires f...
arXiv:2609.36648v1 Announce Type: new Abstract: Vision-language pre-training has reshaped image clustering, giving rise to language-assisted image clustering (LaIC), which leverages textual semantics...
arXiv:2609.36852v1 Announce Type: new Abstract: Trajectory prediction is a key component for understanding human behavior patterns in dynamic scenes. Researchers have devoted substantial efforts to m...
arXiv:2609.36875v1 Announce Type: new Abstract: Accurate instance segmentation in dynamic scenes is important for downstream applications such as robotics and autonomous driving. Existing Segment Any...
arXiv:2609.36891v1 Announce Type: new Abstract: Panoptic segmentation in forest environments is bottlenecked not by semantic quality but by instance separation; existing unsupervised panoptic approac...
arXiv:2609.36906v1 Announce Type: new Abstract: Reliable embodied decisions under partial observability require informative observations and sufficient supporting evidence. However, semantic scores a...
arXiv:2609.36940v1 Announce Type: new Abstract: Accurate dynamic scene reconstruction is important for robotic perception, where temporally consistent representations of dynamic environments are esse...
arXiv:2609.37004v1 Announce Type: new Abstract: We present World2Motion, a framework that generates scene-aware 3D human motion and corresponding video from a single image and a text prompt. While ex...
arXiv:2609.37263v1 Announce Type: new Abstract: While Large Vision-Language Models (LVLMs) achieve remarkable success, hallucinations remain a significant barrier to their reliable deployment. Recent...
arXiv:2609.37386v1 Announce Type: new Abstract: Pre-operative planning for bronchoscopy is critical for the diagnosis of lung lesions. Current accessibility assessment relies on subjective manual ins...
arXiv:2609.37481v1 Announce Type: new Abstract: Large-scale pervasive sensing increasingly relies on high-resolution satellite imagery, yet task-specific onboard vision is constrained by costly annot...
arXiv:2604.02583v3 Announce Type: replace Abstract: We propose FusionBERT, a novel multi-view visual fusion framework for image--3D multimodal retrieval. Existing image--3D representation learning me...
arXiv:2607.28225v2 Announce Type: replace Abstract: Agentic vision-language models (VLMs), which interleave textual reasoning with explicit tool calls such as cropping and code-based image manipulati...
The paper proposes Layer-Informed Fine-Tuning (LIFT), a method that identifies and updates only the most functionally critical layers of large language models (LLMs) using a bottleneck identification mechanism based on sensitivity analysis. By focusing on layers that handle conceptualization, reasoning, and textualization, LIFT aims to accelerate training and enhance performance on reasoning tasks. Experiments demonstrate that this selective fine-tuning approach both speeds up the training process and yields significant performance gains.
ReMem is a new recommendation agent framework that rethinks perception and memory for long-context recommendation tasks. It replaces raw HTML parsing with OCR-based multimodal perception from screenshots, extracting structured information in a platform-agnostic way. The framework also introduces a chunk-wise sequential memory update strategy and a multi-memory GRPO variant to efficiently model evolving user preferences over arbitrarily long interaction histories, achieving a 5.16% average improvement over state-of-the-art baselines on three recommendation agent tasks.
AerialDojo-200K is a large-scale benchmark suite for open-world aerial object-goal search, featuring 42 simulation scenes across four families and 21 types, including urban, natural, infrastructure, and disaster environments. The dataset contains 205,732 task instances—over 100K semantic-goal and over 100K image-goal tasks—each with a collision-free reference trajectory and multi-view video recordings. A unified evaluation framework splits scenes into 21 in-distribution and 21 out-of-distribution sets, and preliminary tests on multimodal large language models show significant room for improvement in general-purpose aerial agents.
The study investigates whether attention heads in large language models that align with human EEG signals are causally involved in model computation. By ablating these brain‑aligned heads during a pattern‑completion task, the authors find that while such heads contribute to performance, their removal is less disruptive than removing heads selected by attribution patching. The research also distinguishes two families of brain‑aligned heads—novelty and repetition heads—highlighting that novelty heads track human attention but are less critical than random ablation, whereas repetition heads modestly aid performance and align with abstract‑pattern representations.
arXiv:2609.37287v1 Announce Type: cross Abstract: Image translation is a fundamental capability of multimodal models for multilingual applications, requiring visual understanding and meaning preserva...