Tootfinder

ARS: Adaptive Reasoning Suppression for Efficient Large Reasoning Language Models
Dongqi Zheng
https://arxiv.org/abs/2510.00071 https://arxiv.org/pdf/2510.…

ARS: Adaptive Reasoning Suppression for Efficient Large Reasoning Language Models
Large Reasoning Language Models (LRLMs or LRMs) demonstrate remarkable capabilities in complex reasoning tasks, but suffer from significant computational inefficiencies due to overthinking phenomena. Existing efficient reasoning methods face the challenge of balancing reasoning quality with inference cost reduction. We propose \textbf{Adaptive Reasoning Suppression (ARS)}, a novel training-free approach that dynamically suppresses redundant reasoning steps while preserving accuracy through adap…

Look Less, Reason More: Rollout-Guided Adaptive Pixel-Space Reasoning
Xuchen Li, Xuzhao Li, Jiahui Gao, Renjie Pi, Shiyu Hu, Wentao Zhang
https://arxiv.org/abs/2510.01681 https:…

Look Less, Reason More: Rollout-Guided Adaptive Pixel-Space Reasoning
Vision-Language Models (VLMs) excel at many multimodal tasks, yet they frequently struggle with tasks requiring precise understanding and handling of fine-grained visual elements. This is mainly due to information loss during image encoding or insufficient attention to critical regions. Recent work has shown promise by incorporating pixel-level visual information into the reasoning process, enabling VLMs to access high-resolution visual details during their thought process. However, this pixel-…

More Than One Teacher: Adaptive Multi-Guidance Policy Optimization for Diverse Exploration
Xiaoyang Yuan, Yujuan Ding, Yi Bin, Wenqi Shao, Jinyu Cai, Jingkuan Song, Yang Yang, Hengtao Shen
https://arxiv.org/abs/2510.02227

More Than One Teacher: Adaptive Multi-Guidance Policy Optimization for Diverse Exploration
Reinforcement Learning with Verifiable Rewards (RLVR) is a promising paradigm for enhancing the reasoning ability in Large Language Models (LLMs). However, prevailing methods primarily rely on self-exploration or a single off-policy teacher to elicit long chain-of-thought (LongCoT) reasoning, which may introduce intrinsic model biases and restrict exploration, ultimately limiting reasoning diversity and performance. Drawing inspiration from multi-teacher strategies in knowledge distillation, we…

Ultra Fast Warm Start Solution for Graph Recommendations
Viacheslav Yusupov, Maxim Rakhuba, Evgeny Frolov
https://arxiv.org/abs/2509.01549 https://arxiv.or…

Ultra Fast Warm Start Solution for Graph Recommendations
In this work, we present a fast and effective Linear approach for updating recommendations in a scalable graph-based recommender system UltraGCN. Solving this task is extremely important to maintain the relevance of the recommendations under the conditions of a large amount of new data and changing user preferences. To address this issue, we adapt the simple yet effective low-rank approximation approach to the graph-based model. Our method delivers instantaneous recommendations that are up to 3…

Resonant production of sterile neutrino dark matter with a refined numerical scheme
Kentaro Kasai, Masahiro Kawasaki, Kai Murai
https://arxiv.org/abs/2510.01907 https://

Resonant production of sterile neutrino dark matter with a refined numerical scheme
The existence of a large primordial neutrino asymmetry is an intriguing possibility, both observationally and theoretically. Such an asymmetry can lead to the resonant production of $\mathrm{keV}$-scale sterile neutrinos, which are a fascinating candidate for dark matter. In this paper, we comprehensively revisit the resonant production processes with a refined numerical analysis, adopting a dynamical discretization of momentum modes to take care of the sharpness of the resonance. We find param…

ACPO: Adaptive Curriculum Policy Optimization for Aligning Vision-Language Models in Complex Reasoning
Yunhao Wang, Ziting Li, Shuai Chen, Tao Liu, Chao Song, Junjie Jiang, Jian Zhu, Peng Gao, Bin Qin
https://arxiv.org/abs/2510.00690

ACPO: Adaptive Curriculum Policy Optimization for Aligning Vision-Language Models in Complex Reasoning
Aligning large-scale vision-language models (VLMs) for complex reasoning via reinforcement learning is often hampered by the limitations of existing policy optimization algorithms, such as static training schedules and the rigid, uniform clipping mechanism in Proximal Policy Optimization (PPO). In this work, we introduce Adaptive Curriculum Policy Optimization (ACPO), a novel framework that addresses these challenges through a dual-component adaptive learning strategy. First, ACPO employs a dyn…

Explore Briefly, Then Decide: Mitigating LLM Overthinking via Cumulative Entropy Regulation
Tianyi Jiang, Yi Bin, Yujuan Ding, Kainian Zhu, Fei Ma, Jingkuan Song, Heng Tao Shen
https://arxiv.org/abs/2510.02249

Explore Briefly, Then Decide: Mitigating LLM Overthinking via Cumulative Entropy Regulation
Large Language Models (LLMs) have demonstrated remarkable reasoning abilities on complex problems using long Chain-of-Thought (CoT) reasoning. However, they often suffer from overthinking, meaning generating unnecessarily lengthy reasoning steps for simpler problems. This issue may degrade the efficiency of the models and make them difficult to adapt the reasoning depth to the complexity of problems. To address this, we introduce a novel metric Token Entropy Cumulative Average (TECA), which mea…

Algorithm Adaptation Bias in Recommendation System Online Experiments
Chen Zheng, Zhenyu Zhao
https://arxiv.org/abs/2509.00199 https://arxiv.org/pdf/2509.0…

Algorithm Adaptation Bias in Recommendation System Online Experiments
Online experiments (A/B tests) are widely regarded as the gold standard for evaluating recommender system variants and guiding launch decisions. However, a variety of biases can distort the results of the experiment and mislead decision-making. An underexplored but critical bias is algorithm adaptation effect. This bias arises from the flywheel dynamics among production models, user data, and training pipelines: new models are evaluated on user data whose distributions are shaped by the incumbe…

Drones that Think on their Feet: Sudden Landing Decisions with Embodied AI
Diego Ortiz Barbosa, Mohit Agrawal, Yash Malegaonkar, Luis Burbano, Axel Andersson, Gy\"orgy D\'an, Henrik Sandberg, Alvaro A. Cardenas
https://arxiv.org/abs/2510.00167

Drones that Think on their Feet: Sudden Landing Decisions with Embodied AI
Autonomous drones must often respond to sudden events, such as alarms, faults, or unexpected changes in their environment, that require immediate and adaptive decision-making. Traditional approaches rely on safety engineers hand-coding large sets of recovery rules, but this strategy cannot anticipate the vast range of real-world contingencies and quickly becomes incomplete. Recent advances in embodied AI, powered by large visual language models, provide commonsense reasoning to assess context a…

LLM4Rec: Large Language Models for Multimodal Generative Recommendation with Causal Debiasing
Bo Ma, Hang Li, ZeHua Hu, XiaoFan Gui, LuYao Liu, Simon Lau
https://arxiv.org/abs/2510.01622

LLM4Rec: Large Language Models for Multimodal Generative Recommendation with Causal Debiasing
Contemporary generative recommendation systems face significant challenges in handling multimodal data, eliminating algorithmic biases, and providing transparent decision-making processes. This paper introduces an enhanced generative recommendation framework that addresses these limitations through five key innovations: multimodal fusion architecture, retrieval-augmented generation mechanisms, causal inference-based debiasing, explainable recommendation generation, and real-time adaptive learni…