Quantization, distillation, pruning and serving work aimed at the same accuracy for less memory, latency and money.
We present Swift-Image, a compact unified model for text-to-image generation, single-image editing, and multi-image editing. Our goal is to explore how far a relatively small visual generator can be pushed through systematic training engineering under a constrained computational budget.
Whether a language model has improved itself is increasingly judged not by mean accuracy but by which individual problems it gains and loses. Tracking these transitions means differencing two noisy estimates, leaving them vulnerable to measurement artifacts.
The paper presents a method for early detection of fraudulent memecoins (rug pulls) on the Solana blockchain, using a dataset of 6.4 million tokens collected over seven months. It shows that most rug pulls occur within an hour of launch and that classic machine learning models, especially Gradient Boosting (XGBoost), can reliably predict them using only the first five minutes of trading data. Cross‑platform data fusion between PumpFun and Raydium further improves detection by reducing domain shift.
The standard objection to full automation is demand-side: if humans earn nothing, who buys the output? This confuses an accounting role with a biological species.
Offline reinforcement learning improves robotic policies using previously collected data without further environment interaction. Yet prevalent diffusion- and flow-matching robot policies lack tractable likelihoods, limiting their use in likelihood-based offline RL post-training.
A lot of prior work addressed key-value (KV) cache selection and compression by sparse attention to enable long-context inference for transformer language models without excessive hardware budgets. We provide a new method for fine-tuning models with sparse attention.
Streaming reconstruction of Free-Viewpoint Videos (FVVs) supports immersive Internet of Things (IoT) services, such as telepresence and digital twin visualization. Existing methods suffer from high per-frame optimization time and large storage footprints, limiting deployment on resource-constrained Edge-IoT devices.
The study evaluates how different graph topologies influence the localization of the epileptogenic zone (EZ) from resting‑state stereo‑EEG recordings in 40 patients. Using a consistent learning model and leave‑one‑patient‑out validation, the authors compare dense graphs, anatomy‑ and geometry‑informed priors, budgeted sparsification, learned sparsification, and a new Region‑Bridge‑c topology. At about 30% edge retention, Region‑Bridge‑c achieves the highest mean PR‑AUC and ROC‑AUC while using roughly 69% fewer edges than a dense graph, indicating that graph construction significantly impacts EZ localization performance.
The paper presents a method for distributing large language model inference across multiple Intel AI PCs by splitting the model into pipeline shards, each pre‑compiled into an OpenVINO graph. Three key techniques—beam_idx Gather to enable GPU optimizations, speculative decoding on stateful models, and interleaved micro‑batching—allow a two‑node Llama 3.1 8B INT4 pipeline to serve two users at 1.79× the throughput of a single‑node model, while a four‑node deployment can run a 70B model that no single PC can hold. The authors provide code, benchmark logs, and reproduction scripts on GitHub.
arXiv:2608. 18147v1 Announce Type: cross Abstract: Adaptive stochastic quantization (ASQ) is a recently introduced quantization approach that optimizes the Mean Squared Error (MSE) for a given input while preserving unbiasedness.
The paper introduces Group‑Calibrated On‑Policy Distillation (GC‑OPD), a method that aligns token‑level teacher guidance with trajectory‑level verifier rewards for long‑context reasoning tasks. GC‑OPD normalizes rewards within rollout groups, uses the signed teacher‑verifier disagreement as a residual, and distributes this residual across tokens via Relative‑Advantage‑Based Credit Assignment (RACA). Experiments on five long‑context benchmarks show that GC‑OPD improves Qwen3‑4B and Qwen3‑8B checkpoints from 29.08/35.12 to 40.47/44.65, outperforming vanilla OPD and demonstrating the effectiveness of group‑relative residual calibration.
The paper introduces DCGCNet, a dual-codebook graph collaborative network that jointly reconstructs ECG signals and classifies atrial fibrillation. It incorporates a local‑global contrastive module for noise‑invariant feature learning and an adaptive codebook vector quantizer to prevent codebook collapse. The model achieves state‑of‑the‑art intra‑dataset performance and consistently attains AUC > 0.98 across seven cross‑dataset settings, even under realistic noisy conditions.
The paper introduces three distance‑based graph autoencoder variants that add structural penalties to the reconstruction loss. All models use a two‑layer Graph Convolutional Network encoder and a Euclidean‑distance decoder, with two node‑level regularizers: a hub penalty based on degree centrality and a penalty based on Natural Community Local Intrinsic Dimensionality (NC‑LID). Experiments on multiple dynamic graph datasets show that incorporating NC‑LID regularization consistently improves reconstruction performance compared to baselines without structural regularization and to the hub‑aware variant.
The paper introduces DART‑SD, a framework for improving multi‑turn tool‑calling agents by respecting the diamond‑shaped topology of task sub‑goals. It models the execution as an Interaction‑State Transition Graph, identifies critical topological breakpoints during rollouts, and uses these to retrieve recovery references. A progressive self‑distillation process then applies localized supervision only on recovery steps, preserving valid reasoning prefixes and enhancing policy diversity.
ADEPT is a reinforcement‑learning framework that first pre‑trains a dexterous policy on a generic object reposing task and then post‑trains downstream policies using this pretrained behavior as a prior. The approach avoids relearning basic skills for each new task, and employs a stable post‑training recipe—behavior‑cloning distillation, critic warm‑up, and conservative on‑policy updates—to preserve the pretrained capabilities. ADEPT’s joint‑space Geometric Fabric mediates between the policy and the robot, enabling zero‑shot sim‑to‑real transfer on a 23‑DoF Kuka‑Allegro and a 29‑DoF Flexiv‑Sharpa, where the robots solve long‑horizon tasks from challenging initial states at human‑level speed.
SIGMA is a reinforcement‑learning framework for traffic signal control that incorporates a large language model to adaptively tune multiple objectives based on natural‑language emergency commands. It uses rotational data augmentation to learn orientation‑invariant policies and an offline‑to‑online training pipeline to ensure stable deployment. Experiments in SUMO on four Kolkata intersections show that SIGMA reduces waiting times, queue lengths, and improves throughput compared to fixed‑time, actuated, and DQN baselines, with ablation studies confirming robustness to component failures and geometric rotations.
The paper introduces a Tsetlin Machine (TM)-based framework for Non‑Intrusive Load Monitoring (NILM) that can run in real time on resource‑constrained microcontrollers. By reformulating NILM as a classification problem, the authors achieve high accuracy—90 % precision and 96 % recall for two appliances, and 77 % precision and 80 % recall for four appliances on the REDD dataset—while keeping the trained model to only 18 KB of flash memory. On an ESP32, the system delivers inference in 0.43 ms, demonstrating its feasibility for privacy‑preserving edge deployment.
WhiteMatter introduces a novel architecture for Transformers that connects every attention layer to representations from all layers of each past token, allowing connection weights to vary across consumer layers and adapt to the source token. The design uses a router to mix the $L$ layer states of each token into $k$ KV channels, which are cached for subsequent tokens; each consumer layer attends to one channel. Experiments show that WhiteMatter outperforms a vanilla Transformer with 50% more layers and maintains most of this advantage even when the KV-cache is compressed by 50%.
UMER is a Unified Multimodal Embedding and Ranking framework that combines contrastive embeddings with Pair‑Aware Discriminative Reasoning to improve universal multimodal retrieval. It replaces item‑wise reflection with pair‑wise comparison of query–candidate pairs, enabling explicit identification of matching and discrepancy evidence. A mutual distillation strategy transfers reliable pairwise preferences between the embedding and ranking components, and UMER achieves state‑of‑the‑art performance on the MMEB‑V2 benchmark while supporting budget‑adjustable inference.
The paper presents an autoencoder-based acoustic anomaly detection system implemented on Intel’s Loihi 2 neuromorphic processor. It achieves high detection performance—0.9959 AUC on a clean ToyADMOS ToyCar benchmark and 0.7990 source AUC on a noisy DCASE 2026 Task 2 ToyCar benchmark—while operating with only 0.0406–0.0426 mJ of dynamic energy per sample, far below CPU and GPU baselines. The system demonstrates that low‑power, on‑chip inference is feasible for persistent machine monitoring.