Tootfinder

Apple-Paper: Warum Reasoning-Modelle wohl nicht denken
Sie brauchen viel Leistung, liefern aber nicht immer bessere Ergebnisse: Large Reasoning Models sollen die KI revolutionieren. Eine Apple-Studie kritisiert das.

Resa: Transparent Reasoning Models via SAEs
Shangshang Wang, Julian Asilis, \"Omer Faruk Akg\"ul, Enes Burak Bilgin, Ollie Liu, Deqing Fu, Willie Neiswanger
https://arxiv.org/abs/2506.09967

Resa: Transparent Reasoning Models via SAEs
How cost-effectively can we elicit strong reasoning in language models by leveraging their underlying representations? We answer this question with Resa, a family of 1.5B reasoning models trained via a novel and efficient sparse autoencoder tuning (SAE-Tuning) procedure. This method first trains an SAE to capture reasoning abilities from a source model, and then uses the trained SAE to guide a standard supervised fine-tuning process to elicit such abilities in a target model, all using verified…

Athena: Enhancing Multimodal Reasoning with Data-efficient Process Reward Models
Shuai Wang, Zhenhua Liu, Jiaheng Wei, Xuanwu Yin, Dong Li, Emad Barsoum
https://arxiv.org/abs/2506.09532

Athena: Enhancing Multimodal Reasoning with Data-efficient Process Reward Models
We present Athena-PRM, a multimodal process reward model (PRM) designed to evaluate the reward score for each step in solving complex reasoning problems. Developing high-performance PRMs typically demands significant time and financial investment, primarily due to the necessity for step-level annotations of reasoning steps. Conventional automated labeling methods, such as Monte Carlo estimation, often produce noisy labels and incur substantial computational costs. To efficiently generate high-q…

23andMe privacy ombudsman recommends company obtains consent for sale of customer data https://therecord.media/23andme-privacy-ombudsman-recommends-consent-sale

Ach ne, das kann doch nicht sein… oder oder… oder doch?! Marketing #Propaganda ist ja nicht neu:
»Ist Reasoning ein #Marketing-Trick? Apple-Forscher zeigen gravierende Probleme von #KI-Modellen:

Ist Reasoning ein Marketing-Trick? Apple-Forscher zeigen gravierende Probleme von KI-Modellen - t3n – digital pioneers
Apple-Forscher:innen haben ein neues Paper veröffentlicht, das die Fertigkeiten von Reasoning-Modellen infrage stellt. Demnach sollen aktuelle Modelle von OpenAI, Google und Anthropic nur in Benchmarks zuverlässig funktionieren, aber an echten Aufgaben grandios scheitern. Immer mehr KI-Unternehmen setzen ihren Fokus auf Reasoning-Modelle, die in der Lage sein sollen, komplexe Gedankengänge zu haben. Doch selbst OpenAI musste […]

Beyond Personalization: Federated Recommendation with Calibration via Low-rank Decomposition
Jundong Chen, Honglei Zhang, Haoxuan Li, Chunxu Zhang, Zhiwei Li, Yidong Li
https://arxiv.org/abs/2506.09525

Beyond Personalization: Federated Recommendation with Calibration via Low-rank Decomposition
Federated recommendation (FR) is a promising paradigm to protect user privacy in recommender systems. Distinct from general federated scenarios, FR inherently needs to preserve client-specific parameters, i.e., user embeddings, for privacy and personalization. However, we empirically find that globally aggregated item embeddings can induce skew in user embeddings, resulting in suboptimal performance. To this end, we theoretically analyze the user embedding skew issue and propose Personalized Fe…

Reasoning as a Resource: Optimizing Fast and Slow Thinking in Code Generation Models
Zongjie Li, Shuai Wang
https://arxiv.org/abs/2506.09396 https://

Reasoning as a Resource: Optimizing Fast and Slow Thinking in Code Generation Models
This position paper proposes a fundamental shift in designing code generation models: treating reasoning depth as a controllable resource. Rather than being an incidental byproduct of prompting, we argue that the trade-off between rapid, direct answers ("fast thinking") and elaborate, chain-of-thought deliberation ("slow thinking") must be explicitly managed. We contend that optimizing reasoning budgets across the entire model lifecycle - from synthetic data creation and benchmarking to real-wo…

Mistral launches its first reasoning models: Magistral Small, on Hugging Face under an Apache 2.0 license, and Magistral Medium, in preview on Mistral's Le Chat (Kyle Wiggers/TechCrunch)
https://techcrunch.com/2025/06/10/mistral-releases-a-pai…

Mistral releases a pair of AI reasoning models | TechCrunch
Like other reasoning models, Magistral works through problems step-by-step for improved consistency and reliability across topics such as math and physics.

Revisiting Graph Projections for Effective Complementary Product Recommendation
Leandro Anghinoni, Pablo Zivic, Jorge Adrian Sanchez
https://arxiv.org/abs/2506.09209

Revisiting Graph Projections for Effective Complementary Product Recommendation
Complementary product recommendation is a powerful strategy to improve customer experience and retail sales. However, recommending the right product is not a simple task because of the noisy and sparse nature of user-item interactions. In this work, we propose a simple yet effective method to predict a list of complementary products given a query item, based on the structure of a directed weighted graph projected from the user-item bipartite graph. We revisit bipartite graph projections for rec…

Causal Sufficiency and Necessity Improves Chain-of-Thought Reasoning
Xiangning Yu, Zhuohan Wang, Linyi Yang, Haoxuan Li, Anjie Liu, Xiao Xue, Jun Wang, Mengyue Yang
https://arxiv.org/abs/2506.09853

Causal Sufficiency and Necessity Improves Chain-of-Thought Reasoning
Chain-of-Thought (CoT) prompting plays an indispensable role in endowing large language models (LLMs) with complex reasoning capabilities. However, CoT currently faces two fundamental challenges: (1) Sufficiency, which ensures that the generated intermediate inference steps comprehensively cover and substantiate the final conclusion; and (2) Necessity, which identifies the inference steps that are truly indispensable for the soundness of the resulting answer. We propose a causal framework that …