LLaMA.cpp releases version b9784 with major optimizations for hexagon-based MM operations, including 32x32 tiled weight repack, improved dyn.quant handling, and unified kernel parameter management. The release includes new binaries for macOS (arm64 and x64), iOS, and multiple Linux architectures with support for Vulkan, ROCm, and OpenVINO.
A fully differentiable emulator of the Atari 2600 VCS is presented, reproducing all 64 ALE games with bit-for-bit accuracy in RAM and screen output. The system enables gradient-based explainable AI by providing a complex, fully known ground truth, with both Julia and JAX implementations validated against a reference emulator and supporting high-throughput training on GPUs.
AdaR enables flexible test-time computing on graphs without parameter changes by using adaptive recurrence. It derives step dependence as a necessary and sufficient condition for convergence and incorporates normalized step information and representation-target relations into recurrent updates, guided by gradient-based supervision signals. Empirical results show AdaR outperforms strong baselines in both inductive and transductive graph learning settings.
Interleaved speech-language models undergo an implicit transcription phase where spoken words become decodable as text tokens in intermediate layers, despite no speech recognition training. Up to 77% of the data shows the spoken word appearing as a top candidate text prediction, followed by a transition to text-based next-word prediction before returning to speech. This behavior is influenced by interleaved training and text LM initialization, and correlates with spoken knowledge performance.
A rental-search assistant with LLM features and multi-market support faced persistent user defects despite 1,553 passing automated tests. Analysis of 252 bug-fix commits showed 44% of fixes occurred at four unseen seams: browser runtime, non-default market, end-to-end flows, and whole-system level. A fix without a seam guard caused a defect to ship twice, highlighting the need for targeted testing at these boundaries.
Fed-CausalDiff introduces a federated causal diffusion framework that enables do-simulation and policy evaluation in decentralized settings. It decomposes latent state evolution into global and local components, allowing decoupled synchronisation to reduce communication cost while maintaining accurate causal inference.
A browser extension enables local image generation using SDXL models via WebGPU, running on the user's GPU without external setups. The tool supports two models: SDXL-Lighting fp16 (7 GB) and a 4-bit version (3.6 GB), with requirements including at least 8 GB VRAM for the full model and a browser with WebGPU support (Chrome/Edge 122+ or latest Firefox).
A two-stage deep learning model classifies Indian sign language video clips into English words using a fine-tuned VideoMAE transformer, achieving 99% training and 78% validation accuracy on a 13-class dataset. The predicted English labels are translated into Hindi, Telugu, and Bengali using Meta AI's NLLB-200 multilingual model, with a Streamlit demo enabling user-uploaded video inference and cross-lingual output.
A structured prompt framework improves local LLM accuracy in environmental monitoring by transforming raw sensor data into enriched textual representations. Evaluations on indoor and outdoor datasets show local model accuracy increases from 50.9% to 81.7% indoors and from 63.7% to 89.3% outdoors with enriched prompts, while maintaining low latency near 0.22 seconds in no-chain-of-thought mode.
Input diversity, a common practice in transfer attacks, improves success on standard surrogates but reduces it on robust ones. This regime-dependent effect, called the Scissors Effect, is driven by gradient geometry, with resize operations degrading alignment in robust models. A training-free rule (CG-DI) adjusts diversity based on local gradient consistency to preserve attack success across surrogate types.
HERTA is the first automated testing tool designed for fully homomorphic encryption frameworks. It uses metamorphic testing with novel relations derived from FHE semantics to detect deep-seated logic bugs that can silently corrupt encrypted computations. Evaluation on three industry frameworks revealed 21 previously unknown bugs, several of which have been confirmed and fixed by developers, with significant implications for security and service integrity.
A new training method replaces MSE loss in diffusion models with an f-divergence-based transformation, creating a robust surrogate that improves performance under data contamination. The approach uses local divergence constructions under DDPM's Gaussian reverse-kernel, reducing the training objective to a one-dimensional function of denoising error, with bounded-influence divergences suppressing large errors and enhancing stability.
Generative Robust Optimisation (GRO) introduces a deep generative model to define uncertainty sets, capturing nonlinear correlations, asymmetry, and multimodality. A five-point evaluation framework assesses neural network-based uncertainty sets across reconstruction fidelity, distribution matching, latent regularity, robust relevance, and computational tractability, with experiments validating GRO's effectiveness in production planning and facility location problems.
A learnable scaling parameter called Quantum Measurement Temperature (QMT) is introduced to rescale quantum measurement outputs in hybrid quantum neural networks. This approach mitigates measurement-induced logit contraction, enhancing gradient magnitude and stability during training without altering the quantum circuit or measurement operators. Experiments show improved logit separation, gradient strength, and classification accuracy in protein and image classification tasks.
A piezoelectric deep material network (PDMN) is proposed to efficiently homogenize two-phase piezoelectric composites. The framework embeds electromechanical homogenization relations into its architecture, enabling physics-informed, semi-analytical predictions with over three orders of magnitude lower computational cost than direct numerical simulation, validated on PVDF-LiNbO3 and viscoelastic-piezoelectric composites under nonlinear loading.
CCPL introduces a lightweight framework that anchors class prompts to frozen concept prototypes, improving few-shot CLIP adaptation. It achieves better base-to-new performance on DTD and EuroSAT compared to CoOp, with consistent gains from text-space concept regularization, though results vary by dataset and protocol.
This paper introduces a stationary mean-field game framework that directly incorporates distributional model uncertainty into population-coupled dynamics. It establishes a robust dynamic programming principle, proves existence of a stationary robust equilibrium, and presents the first algorithm with convergence guarantees. The mean-field solution approximates finite-population equilibria and provides explicit non-asymptotic error bounds under model uncertainty.
SiM enables dynamic routing in multi-task model merging without additional training or task ID access. It uses SVD-based manifold approximations and projects test inputs onto precomputed task manifolds to route inputs to relevant experts, improving performance and reducing the gap to individual expert levels.
On-Policy Distillation (OPD) suffers from position bias where later tokens provide poor supervision. We introduce Importance-Weighted On-Policy Distillation (IW-OPD), which assigns weights based on distribution discrepancy, prioritizing early tokens. IW-OPD converges faster and achieves up to 6.9 point performance gains on AIME-2025.
A generative surrogate framework using a Variational Autoencoder approximates Gaussian Process priors, bypassing costly covariance operations. The VAE+Hidden Markov Model architecture enables fast, scalable stellar flare detection in large astronomical time series, matching exact models in structural fidelity while reducing computational time significantly.