Category: News

Balancing Rewards in Text Summarization: Multi-Objective Reinforcement Learning via HyperVolume Optimizationcs.AI updates on arXiv.org arXiv:2510.19325v1 Announce Type: cross
Abstract: Text summarization is a crucial task that requires the simultaneous optimization of multiple objectives, including consistency, coherence, relevance, and fluency, which presents considerable challenges. Although large language models (LLMs) have demonstrated remarkable performance, enhanced by reinforcement learning (RL), few studies have focused on optimizing the multi-objective problem of summarization through RL based on LLMs. In this paper, we introduce hypervolume optimization (HVO), a novel optimization strategy that dynamically adjusts the scores between groups during the reward process in RL by using the hypervolume method. This method guides the model’s optimization to progressively approximate the pareto front, thereby generating balanced summaries across multiple objectives. Experimental results on several representative summarization datasets demonstrate that our method outperforms group relative policy optimization (GRPO) in overall scores and shows more balanced performance across different dimensions. Moreover, a 7B foundation model enhanced by HVO performs comparably to GPT-4 in the summarization task, while maintaining a shorter generation length. Our code is publicly available at https://github.com/ai4business-LiAuto/HVO.git

 arXiv:2510.19325v1 Announce Type: cross
Abstract: Text summarization is a crucial task that requires the simultaneous optimization of multiple objectives, including consistency, coherence, relevance, and fluency, which presents considerable challenges. Although large language models (LLMs) have demonstrated remarkable performance, enhanced by reinforcement learning (RL), few studies have focused on optimizing the multi-objective problem of summarization through RL based on LLMs. In this paper, we introduce hypervolume optimization (HVO), a novel optimization strategy that dynamically adjusts the scores between groups during the reward process in RL by using the hypervolume method. This method guides the model’s optimization to progressively approximate the pareto front, thereby generating balanced summaries across multiple objectives. Experimental results on several representative summarization datasets demonstrate that our method outperforms group relative policy optimization (GRPO) in overall scores and shows more balanced performance across different dimensions. Moreover, a 7B foundation model enhanced by HVO performs comparably to GPT-4 in the summarization task, while maintaining a shorter generation length. Our code is publicly available at https://github.com/ai4business-LiAuto/HVO.git Read More  

Study of Training Dynamics for Memory-Constrained Fine-Tuningcs.AI updates on arXiv.org arXiv:2510.19675v1 Announce Type: cross
Abstract: Memory-efficient training of deep neural networks has become increasingly important as models grow larger while deployment environments impose strict resource constraints. We propose TraDy, a novel transfer learning scheme leveraging two key insights: layer importance for updates is architecture-dependent and determinable a priori, while dynamic stochastic channel selection provides superior gradient approximation compared to static approaches. We introduce a dynamic channel selection approach that stochastically resamples channels between epochs within preselected layers. Extensive experiments demonstrate TraDy achieves state-of-the-art performance across various downstream tasks and architectures while maintaining strict memory constraints, achieving up to 99% activation sparsity, 95% weight derivative sparsity, and 97% reduction in FLOPs for weight derivative computation.

 arXiv:2510.19675v1 Announce Type: cross
Abstract: Memory-efficient training of deep neural networks has become increasingly important as models grow larger while deployment environments impose strict resource constraints. We propose TraDy, a novel transfer learning scheme leveraging two key insights: layer importance for updates is architecture-dependent and determinable a priori, while dynamic stochastic channel selection provides superior gradient approximation compared to static approaches. We introduce a dynamic channel selection approach that stochastically resamples channels between epochs within preselected layers. Extensive experiments demonstrate TraDy achieves state-of-the-art performance across various downstream tasks and architectures while maintaining strict memory constraints, achieving up to 99% activation sparsity, 95% weight derivative sparsity, and 97% reduction in FLOPs for weight derivative computation. Read More  

How do AI ‘humanisers’ compare to human editing?AI News The emergence of artificial intelligence has fundamentally altered the field of content creation. Tools capable of generating coherent, often impressive, text, are now ubiquitous. Yet, despite their sophistication, AI-generated content presents a persistent challenge because it often has a “robotic” quality, lacking the warmth, nuance, and genuine voice that connects with a human audience. The
The post How do AI ‘humanisers’ compare to human editing? appeared first on AI News.

 The emergence of artificial intelligence has fundamentally altered the field of content creation. Tools capable of generating coherent, often impressive, text, are now ubiquitous. Yet, despite their sophistication, AI-generated content presents a persistent challenge because it often has a “robotic” quality, lacking the warmth, nuance, and genuine voice that connects with a human audience. The
The post How do AI ‘humanisers’ compare to human editing? appeared first on AI News. Read More  

Top 5 Open Source Video Generation ModelsKDnuggets Discover the top open source video generation models that rival Veo 3 and prioritize your privacy and control.

 Discover the top open source video generation models that rival Veo 3 and prioritize your privacy and control. Read More  

Multiple Linear Regression Explained Simply (Part 1)Towards Data Science The math behind fitting a plane instead of a line.
The post Multiple Linear Regression Explained Simply (Part 1) appeared first on Towards Data Science.

 The math behind fitting a plane instead of a line.
The post Multiple Linear Regression Explained Simply (Part 1) appeared first on Towards Data Science. Read More  

OpenAI data residency advances enterprise AI governanceAI News For chief data and information officers, especially in tightly regulated sectors, data governance has been a major cause preventing enterprise adoption of AI models. The issue of data sovereignty – which concerns where company data is handled and kept – has held many back, forcing them to use complex private cloud solutions. Others have simply
The post OpenAI data residency advances enterprise AI governance appeared first on AI News.

 For chief data and information officers, especially in tightly regulated sectors, data governance has been a major cause preventing enterprise adoption of AI models. The issue of data sovereignty – which concerns where company data is handled and kept – has held many back, forcing them to use complex private cloud solutions. Others have simply
The post OpenAI data residency advances enterprise AI governance appeared first on AI News. Read More  

Insights into the Unknown: Federated Data Diversity Analysis on Molecular Datacs.AI updates on arXiv.org arXiv:2510.19535v1 Announce Type: cross
Abstract: AI methods are increasingly shaping pharmaceutical drug discovery. However, their translation to industrial applications remains limited due to their reliance on public datasets, lacking scale and diversity of proprietary pharmaceutical data. Federated learning (FL) offers a promising approach to integrate private data into privacy-preserving, collaborative model training across data silos. This federated data access complicates important data-centric tasks such as estimating dataset diversity, performing informed data splits, and understanding the structure of the combined chemical space. To address this gap, we investigate how well federated clustering methods can disentangle and represent distributed molecular data. We benchmark three approaches, Federated kMeans (Fed-kMeans), Federated Principal Component Analysis combined with Fed-kMeans (Fed-PCA+Fed-kMeans), and Federated Locality-Sensitive Hashing (Fed-LSH), against their centralized counterparts on eight diverse molecular datasets. Our evaluation utilizes both, standard mathematical and a chemistry-informed evaluation metrics, SF-ICF, that we introduce in this work. The large-scale benchmarking combined with an in-depth explainability analysis shows the importance of incorporating domain knowledge through chemistry-informed metrics, and on-client explainability analyses for federated diversity analysis on molecular data.

 arXiv:2510.19535v1 Announce Type: cross
Abstract: AI methods are increasingly shaping pharmaceutical drug discovery. However, their translation to industrial applications remains limited due to their reliance on public datasets, lacking scale and diversity of proprietary pharmaceutical data. Federated learning (FL) offers a promising approach to integrate private data into privacy-preserving, collaborative model training across data silos. This federated data access complicates important data-centric tasks such as estimating dataset diversity, performing informed data splits, and understanding the structure of the combined chemical space. To address this gap, we investigate how well federated clustering methods can disentangle and represent distributed molecular data. We benchmark three approaches, Federated kMeans (Fed-kMeans), Federated Principal Component Analysis combined with Fed-kMeans (Fed-PCA+Fed-kMeans), and Federated Locality-Sensitive Hashing (Fed-LSH), against their centralized counterparts on eight diverse molecular datasets. Our evaluation utilizes both, standard mathematical and a chemistry-informed evaluation metrics, SF-ICF, that we introduce in this work. The large-scale benchmarking combined with an in-depth explainability analysis shows the importance of incorporating domain knowledge through chemistry-informed metrics, and on-client explainability analyses for federated diversity analysis on molecular data. Read More  

Style Attack Disguise: When Fonts Become a Camouflage for Adversarial Intentcs.AI updates on arXiv.org arXiv:2510.19641v1 Announce Type: cross
Abstract: With social media growth, users employ stylistic fonts and font-like emoji to express individuality, creating visually appealing text that remains human-readable. However, these fonts introduce hidden vulnerabilities in NLP models: while humans easily read stylistic text, models process these characters as distinct tokens, causing interference. We identify this human-model perception gap and propose a style-based attack, Style Attack Disguise (SAD). We design two sizes: light for query efficiency and strong for superior attack performance. Experiments on sentiment classification and machine translation across traditional models, LLMs, and commercial services demonstrate SAD’s strong attack performance. We also show SAD’s potential threats to multimodal tasks including text-to-image and text-to-speech generation.

 arXiv:2510.19641v1 Announce Type: cross
Abstract: With social media growth, users employ stylistic fonts and font-like emoji to express individuality, creating visually appealing text that remains human-readable. However, these fonts introduce hidden vulnerabilities in NLP models: while humans easily read stylistic text, models process these characters as distinct tokens, causing interference. We identify this human-model perception gap and propose a style-based attack, Style Attack Disguise (SAD). We design two sizes: light for query efficiency and strong for superior attack performance. Experiments on sentiment classification and machine translation across traditional models, LLMs, and commercial services demonstrate SAD’s strong attack performance. We also show SAD’s potential threats to multimodal tasks including text-to-image and text-to-speech generation. Read More  

ToMMeR — Efficient Entity Mention Detection from Large Language Modelscs.AI updates on arXiv.org arXiv:2510.19410v1 Announce Type: cross
Abstract: Identifying which text spans refer to entities — mention detection — is both foundational for information extraction and a known performance bottleneck. We introduce ToMMeR, a lightweight model (<300K parameters) probing mention detection capabilities from early LLM layers. Across 13 NER benchmarks, ToMMeR achieves 93% recall zero-shot, with over 90% precision using an LLM as a judge showing that ToMMeR rarely produces spurious predictions despite high recall. Cross-model analysis reveals that diverse architectures (14M-15B parameters) converge on similar mention boundaries (DICE >75%), confirming that mention detection emerges naturally from language modeling. When extended with span classification heads, ToMMeR achieves near SOTA NER performance (80-87% F1 on standard benchmarks). Our work provides evidence that structured entity representations exist in early transformer layers and can be efficiently recovered with minimal parameters.

 arXiv:2510.19410v1 Announce Type: cross
Abstract: Identifying which text spans refer to entities — mention detection — is both foundational for information extraction and a known performance bottleneck. We introduce ToMMeR, a lightweight model (<300K parameters) probing mention detection capabilities from early LLM layers. Across 13 NER benchmarks, ToMMeR achieves 93% recall zero-shot, with over 90% precision using an LLM as a judge showing that ToMMeR rarely produces spurious predictions despite high recall. Cross-model analysis reveals that diverse architectures (14M-15B parameters) converge on similar mention boundaries (DICE >75%), confirming that mention detection emerges naturally from language modeling. When extended with span classification heads, ToMMeR achieves near SOTA NER performance (80-87% F1 on standard benchmarks). Our work provides evidence that structured entity representations exist in early transformer layers and can be efficiently recovered with minimal parameters. Read More  

Graph Unlearning Meets Influence-aware Negative Preference Optimizationcs.AI updates on arXiv.org arXiv:2510.19479v1 Announce Type: cross
Abstract: Recent advancements in graph unlearning models have enhanced model utility by preserving the node representation essentially invariant, while using gradient ascent on the forget set to achieve unlearning. However, this approach causes a drastic degradation in model utility during the unlearning process due to the rapid divergence speed of gradient ascent. In this paper, we introduce textbf{INPO}, an textbf{I}nfluence-aware textbf{N}egative textbf{P}reference textbf{O}ptimization framework that focuses on slowing the divergence speed and improving the robustness of the model utility to the unlearning process. Specifically, we first analyze that NPO has slower divergence speed and theoretically propose that unlearning high-influence edges can reduce impact of unlearning. We design an influence-aware message function to amplify the influence of unlearned edges and mitigate the tight topological coupling between the forget set and the retain set. The influence of each edge is quickly estimated by a removal-based method. Additionally, we propose a topological entropy loss from the perspective of topology to avoid excessive information loss in the local structure during unlearning. Extensive experiments conducted on five real-world datasets demonstrate that INPO-based model achieves state-of-the-art performance on all forget quality metrics while maintaining the model’s utility. Codes are available at href{https://github.com/sh-qiangchen/INPO}{https://github.com/sh-qiangchen/INPO}.

 arXiv:2510.19479v1 Announce Type: cross
Abstract: Recent advancements in graph unlearning models have enhanced model utility by preserving the node representation essentially invariant, while using gradient ascent on the forget set to achieve unlearning. However, this approach causes a drastic degradation in model utility during the unlearning process due to the rapid divergence speed of gradient ascent. In this paper, we introduce textbf{INPO}, an textbf{I}nfluence-aware textbf{N}egative textbf{P}reference textbf{O}ptimization framework that focuses on slowing the divergence speed and improving the robustness of the model utility to the unlearning process. Specifically, we first analyze that NPO has slower divergence speed and theoretically propose that unlearning high-influence edges can reduce impact of unlearning. We design an influence-aware message function to amplify the influence of unlearned edges and mitigate the tight topological coupling between the forget set and the retain set. The influence of each edge is quickly estimated by a removal-based method. Additionally, we propose a topological entropy loss from the perspective of topology to avoid excessive information loss in the local structure during unlearning. Extensive experiments conducted on five real-world datasets demonstrate that INPO-based model achieves state-of-the-art performance on all forget quality metrics while maintaining the model’s utility. Codes are available at href{https://github.com/sh-qiangchen/INPO}{https://github.com/sh-qiangchen/INPO}. Read More