The paper introduces a curvature‑aware Expected Free Energy (EFE) acquisition function for Bayesian optimization, designed to jointly learn and optimize an underlying function. It demonstrates that, under certain assumptions, EFE reduces to familiar criteria such as Upper Confidence Bound, Lower Confidence Bound, and Expected Information Gain, and provides unbiased convergence guarantees for concave functions. Empirical results on a Van der Pol oscillator system identification task and a two‑dimensional oscillatory benchmark show that the adaptive EFE achieves competitive performance in both regret and mean squared error, outperforming typical acquisition functions that excel in only one metric.
By Ajith Anil Meera, Wouter Kouw
ASPIRE introduces a non‑synchronized batched self‑speculative decoding framework for long‑context LLM inference, addressing the memory bottleneck of attention by drafting tokens with sparse attention and verifying them with full attention. It combines a unified mixed forward pass, a lightweight online speculation scheduler that lets each request independently decide when to verify, and an intra‑draft refresh layer that updates the sparse context at every draft step. Experiments on three models and five benchmarks show 1.70–4.58× speedup over autoregressive baselines and a 27% average improvement over the strongest prior self‑speculative methods.
By Amir Ziashahabi, Hossein Entezari Zarch, Lei Gao, Murali Annavaram, Salman Avestimehr
arXiv:2609.17981v1 Announce Type: cross
Abstract: Speech Large Language Models (Speech-LLMs), typically built from a pre-trained speech encoder, a modality projector, and an LLM fine-tuned with Low-R...
By Mohan Shi, Zilai Wang, Natarajan Balaji Shankar, Kaiyuan Zhang, Eray Eren, Abeer Alwan
The paper presents a reinforcement‑learning framework that dynamically selects between full‑order models (FOMs) and reduced‑order models (ROMs) in hybrid domain decomposition simulations using the overlapping Schwarz alternating method (O‑SAM). Offline‑trained Deep Q‑networks choose subdomain‑local FOMs or pre‑trained Operator Inference ROMs based on a reward that balances accuracy, computational cost, and model‑switching frequency, and the learned policies are deployed on unseen problem instances without needing a reference FOM solution. Experiments on a 1D advection‑diffusion problem and a 3D elastic wave propagation benchmark show that the RL‑guided policies adaptively allocate high‑fidelity resolution as features propagate, outperforming static FOM/ROM assignments and demonstrating the feasibility of predictive online fidelity adaptation in Schwarz‑based hybrid simulations.
By Trishit Mondal, Irina Tezaur, Anthony Gruber
arXiv:2609.19142v1 Announce Type: new
Abstract: World models endow perceptual systems with the ability to predict how scenes evolve under interaction. They are most beneficial when trained on diverse...
By Bardienus P. Duisterhof, Kaifeng Zhang, Adam Hung, Bowen Wen, Stan Birchfield, Yunzhu Li, Deva Ramanan, Jeffrey Ichnowski
The paper studies Graph-Guided Token Merging (G2TM), a module that reduces token count in Vision Transformers. It evaluates G2TM across multiple segmentation frameworks and decoder types, finding that its performance gains are tied to the encoder rather than the decoder. The authors report consistent reductions in GFLOPs (22‑47%) and throughput improvements (up to 74%) on ADE20K, with optimal hyperparameters depending mainly on backbone pre‑training and target dataset.
By Victor Bercy, Martyna Poreba, Michal Szczepanski, Samia Bouchafa
The paper introduces a method for offline on‑policy distillation that addresses the problem of imperfect teacher supervision. By training on teacher‑successful problems and measuring changes in token likelihoods on teacher‑failed trajectories, the authors derive a learnability signal that weights the distillation loss. This approach improves performance on mathematical reasoning and code generation tasks while reducing computational cost compared to online distillation.
By Yihao Ai, Weilong Yan
This empirical study is an independent reproduction of the dissociation Zhao reported in 2026. The shape of a large language model's chain-of-thought entropy trajectory predicts whether the final answ...
Artificial intelligence for plant disease analysis has advanced from task-specific classifiers to multi-modal models capable of jointly interpreting visual and textual information. However, practical...
Counterfactual (CF) explanations identify changes that alter an input's classification. While existing methods produce realistic and low-cost CFs, they often fail to ensure feasibility, by suggesting...
Intelligent systems that act in the world require image understanding that is both comprehensive and spatially grounded. Current vision-language models (VLMs) can generate fluent and detailed image ca...
World models endow perceptual systems with the ability to predict how scenes evolve under interaction. They are most beneficial when trained on diverse volumes of data, to instill a rich prior into do...
As models scale, reward hacking becomes more frequent, more sophisticated, and more consequential. Does it leave a telltale signature in model representations? This work analyzes how reward hacking is...
Reinforcement learning (RL) is an exciting concept as well as a remarkable success story worth sharing. However, RL builds on rather complex interactions between different objects that play out over s...
Mixture-of-Experts (MoE) language models suffer from large parameter counts, which create a significant memory bottleneck. Expert pruning is the most direct approach for reducing this parameter count,...
Existing legal benchmarks cover diverse tasks, while continuous monetary remedies remain comparatively underexplored. We introduce ECtHR-NPD, to the best of our knowledge, the first benchmark for pred...
Current Mixture-of-Agents (MoA) paradigms generally treat query routing and agent fine-tuning as separate processes, limiting their ability to respond to evolving agent capabilities. This disconnect p...
Correctly answering a question grounded in normative documents often depends on information outside any single passage: whether the retrieved document is the version currently in force; whether it app...
Multi-modal talking avatar synthesis aims to generate realistic talking videos from a reference portrait and speech. Despite rapid progress in diffusion-based methods, existing approaches still strugg...
Most keyframe selection studies focus on offline settings, assuming access to the full video and query in advance. In contrast, real-world streaming scenarios require online frame selection under unkn...