arXiv:2605.05054v1 Announce Type: cross Abstract: Recent flow matching (FM) methods improve the few-shot adaptation of vision-language models, by modeling cross-modal alignment as a continuous multi-step flow. In this paper, we argue that existing FM methods are inherently constrained by incompatible geometric priors on pre-trained cross-modal features, resulting in suboptimal adaptation performance. We first analyze these […]
Neuro-Symbolic Agents for Hallucination-Free Requirements Reuse
arXiv:2605.01562v2 Announce Type: replace-cross Abstract: The Object-Oriented Method for Requirements Authoring and Management (OOMRAM) is a requirements reuse framework that relies on exact identifier matching and rigid templates, limiting its ability to adapt specifications across diverse contexts. While Large Language Models (LLMs) offer the flexibility to overcome this bottleneck, they introduce the risk of generating […]
When LLMs get significantly worse: A statistical approach to detect model degradations
arXiv:2602.10144v2 Announce Type: replace-cross Abstract: Minimizing the inference cost and latency of foundation models has become a crucial area of research. Optimization approaches include theoretically lossless methods and others without accuracy guarantees like quantization. In all of these cases it is crucial to ensure that the model quality has not degraded. However, even at temperature […]
The Tsetlin Machine Goes Deep: Logical Learning and Reasoning With Graphs
arXiv:2507.14874v2 Announce Type: replace-cross Abstract: Pattern recognition with concise and flat AND-rules makes the Tsetlin Machine (TM) both interpretable and efficient, while the power of Tsetlin automata enables accuracy comparable to deep learning on an increasing number of datasets. We introduce the Graph Tsetlin Machine (GraphTM) for learning interpretable deep clauses from graph-structured input. Moving […]
Fragile Knowledge, Robust Instruction-Following: The Width Pruning Dichotomy in Llama-3.2
arXiv:2512.22671v2 Announce Type: replace-cross Abstract: Structured width pruning of GLU-MLP layers, guided by the Maximum Absolute Weight (MAW) criterion, reveals a systematic dichotomy in how reducing the expansion ratio affects different model capabilities. While performance on tasks relying on parametric knowledge (e.g., MMLU, GSM8K) and perplexity metrics degrades predictably, instruction-following capabilities improve substantially (+46% to […]
Governed Capability Evolution for Embodied Agents: Safe Upgrade, Compatibility Checking, and Runtime Rollback for Embodied Capability Modules
arXiv:2604.08059v3 Announce Type: replace-cross Abstract: Embodied agents are increasingly expected to improve over time by updating their executable capabilities rather than rewriting the agent itself. Prior work has separately studied modular capability packaging, capability evolution, and runtime governance. However, a key systems problem remains underexplored: once an embodied capability module evolves into a new version, […]
Taming Outlier Tokens in Diffusion Transformers
arXiv:2605.05206v1 Announce Type: cross Abstract: We study outlier tokens in Diffusion Transformers (DiTs) for image generation. Prior work has shown that Vision Transformers (ViTs) can produce a small number of high-norm tokens that attract disproportionate attention while carrying limited local information, but their role in generative models remains underexplored. We show that this phenomenon appears […]
Multiple timescales in collective motion: daily and intraday upstream fish migration focusing on Feller condition
arXiv:2602.06606v2 Announce Type: replace Abstract: Fish migration is a collective phenomenon that has multiple timescales, ranging from daily to intraday (hourly or even finer). We propose a unified mathematical approach using diffusion bridges, nonlinear stochastic differential equations with pinned initial and terminal conditions, to model both daily and intraday fish migration phenomena. Drift and diffusion […]
Valley3: Scaling Omni Foundation Models for E-commerce
arXiv:2605.01278v2 Announce Type: replace Abstract: In this work, we present Valley3, an omni multimodal large language model (MLLM) developed for diverse global e-commerce tasks, with unified understanding and reasoning capabilities across text, images, video, and audio. A key feature of Valley3 is its native multilingual audio capability for e-commerce, developed by extending vision-language models to […]
Dataset-Driven Channel Masks in Transformers for Multivariate Time Series
arXiv:2410.23222v3 Announce Type: replace-cross Abstract: Recent advancements in foundation models have been successfully extended to the time series (TS) domain, facilitated by the emergence of large-scale TS datasets. However, previous efforts have primarily Capturing channel dependency (CD) is essential for modeling multivariate time series (TS), and attention-based methods have been widely employed for this purpose. […]
Diffusion-Inspired Masked Fine-Tuning for Knowledge Injection in Autoregressive LLMs
arXiv:2510.09885v5 Announce Type: replace-cross Abstract: Large language models (LLMs) are often used in environments where facts evolve, yet factual knowledge updates via fine-tuning on unstructured text often suffer from 1) reliance on compute-heavy paraphrasing augmentation and 2) the reversal curse. Recent studies show diffusion large language models (dLLMs) require fewer training samples to achieve lower […]
Syntax- and Compilation-Preserving Evasion of LLM Vulnerability Detectors
arXiv:2602.00305v2 Announce Type: replace-cross Abstract: LLM-based vulnerability detectors are increasingly deployed in CI/CD security gating, yet their resilience to evasion under syntax- and compilation-preserving edits remains poorly understood. We evaluate five attack variants spanning four carrier families of behavior-preserving code transformations on a unified C/C++ benchmark ($N=5000$) and introduce Complete Resistance (CR), measuring the fraction […]