Quantization, distillation, pruning and serving work aimed at the same accuracy for less memory, latency and money.
arXiv:2608.21500v1 Announce Type: cross Abstract: Prompt injection is listed as the \#1 threat to AI agents. When an agent accesses external data from websites, files, or emails, an attacker may inje...
arXiv:2608.21487v1 Announce Type: cross Abstract: Vision-Language Models (VLMs) exhibit strong zero-shot capabilities, making them an attractive solution for continual learning across diverse tasks....
arXiv:2608.23012v1 Announce Type: new Abstract: Image matching is a core component of applications such as Simultaneous Localization and Mapping (SLAM), Visual Localization, and Structure from Motion...
arXiv:2608.22054v1 Announce Type: new Abstract: Structure-from-Motion (SfM) aims to estimate camera poses and reconstruct 3D structures from a collection of unordered images. Compared with incrementa...
arXiv:2602.13106v2 Announce Type: replace-cross Abstract: In recent years, there has been growing interest in understanding neural architectures' ability to learn to execute discrete algorithms, a li...
arXiv:2608.21764v1 Announce Type: cross Abstract: Event-based vision has emerged as a promising paradigm for energy-aware artificial intelligence (AI), offering sparse, low-latency visual signals tha...
arXiv:2608.22979v1 Announce Type: new Abstract: Deploying high-dimensional multimodal features in industrial recommender systems incurs substantial storage and latency overhead. Hard quantization is...
arXiv:2608.21450v1 Announce Type: new Abstract: Knowledge-Based Visual Question Answering (KB-VQA) relies on retrieving external information to answer queries involving long-tail entities. However, e...
arXiv:2608.21946v1 Announce Type: cross Abstract: Reinforcement learning with outcome-based objectives such as GRPO enables LLM-based agents to solve complex, long-horizon tasks, yet the reusable exp...
arXiv:2608.22704v1 Announce Type: new Abstract: Long-form audio inputs make the KV cache the dominant memory cost of speech LLMs. Prefill-only KV compression methods permanently discard audio KV posi...
arXiv:2608.22917v1 Announce Type: new Abstract: We present TSWAP, a deployed eight-language conversational wellness advisor grounded, via retrieval-augmented generation, in a verified knowledge base...
XTC (Exclude Top Choices) is a lightweight, head‑aware decoding operator that improves diversity in autoregressive language models by removing overly probable tokens that dominate the next‑token distribution. It works by identifying tokens above a plausibility threshold, probabilistically excluding the dominant choices, and renormalizing the remaining distribution. Across 60 experiments on models such as Gemma 3 and DeepSeek R1, XTC boosts Distinct‑2 scores by 11–15 % and cuts repeat trigrams by 27–47 %, while a Mechanical Turk study shows a 62.3 % preference for XTC‑generated text without loss of fluency.
arXiv:2608.23253v1 Announce Type: cross Abstract: Vision-language models typically encode an image into hundreds of visual tokens, incurring substantial inference latency and GPU memory overhead. Exi...
arXiv:2608.23101v1 Announce Type: cross Abstract: Recent progress in the field of TinyML has demonstrated that low-power hardware based on microcontrollers can achieve bird species monitoring in real...
arXiv:2608.21445v1 Announce Type: new Abstract: Automated seizure detection from electroencephalography (EEG) is essential for continuous neurological monitoring, particularly for subclinical epilept...
The paper introduces a training‑free structured suffix modeling technique to accelerate Diffusion Language Models (DLMs). It partitions the suffix into local, middle, and tail regions, retaining varying numbers of tokens per region and incorporating previous decoding results into current token representations. Experiments on three DLMs show significant speedups—up to 72.81× in long‑sequence inference—while often improving performance, and the method is compatible with existing acceleration strategies.
The paper introduces CloudEdgeVLA, a cloud‑edge policy for Vision‑Language‑Action models that treats temporal misalignment as a representation‑learning problem. It encodes delayed observations into slowly varying task features on the cloud while a lightweight edge head fuses the latest cloud feature with current local vision. Experiments on four LIBERO suites show that CloudEdgeVLA retains 63.8–78.0% success under a 40‑step delay window, far outperforming VLASH and single‑path baselines.
The paper revisits the impact of pruning on large language models (LLMs) during test-time scaling (TTS). While prior work found that structured pruning degrades reasoning performance, this study shows that unstructured pruning—removing only specific redundant weights—can actually improve TTS performance on reasoning benchmarks for models s1.1-7B and Qwen3-8B, sometimes surpassing the full-weight models. The authors also examine how different layer-wise sparsity allocation strategies affect these outcomes.
The paper introduces Align‑LoRA, a unified LoRA framework for multi‑task learning that replaces complex, isolated adapter designs with a single‑adapter model enhanced by a higher rank and an explicit alignment loss. It demonstrates that a router‑free, multi‑head model with high inter‑head redundancy can outperform more elaborate baselines, and that a unified LoRA can achieve competitive performance while enabling weight merging and zero inference latency. Extensive experiments and theoretical analysis confirm that Align‑LoRA surpasses prevailing approaches, offering a simpler, production‑friendly paradigm for parameter‑efficient fine‑tuning of large language models.
MOCLIP (Metasurface Optics Contrastive Learning Pretrained) is a nanophotonic foundation model that unifies metasurface geometry and spectra in a shared latent space using contrastive learning on an ImageNet‑1K–sized experimental dataset. It enables high‑throughput zero‑shot inverse design, predicting 0.2 million samples per second and allowing the design of a full 4‑inch wafer of high‑density metasurfaces in minutes. The model also supports generative latent‑space optimization with 97 % accuracy and demonstrates an optical information storage concept achieving 0.1 Gbit/mm², six times higher than commercial optical media.