Tootfinder

Explaining multimodal LLMs via intra-modal token interactions
Jiawei Liang, Ruoyu Chen, Xianghao Jiao, Siyuan Liang, Shiming Liu, Qunli Zhang, Zheng Hu, Xiaochun Cao
https://arxiv.org/abs/2509.22415

Explaining multimodal LLMs via intra-modal token interactions
Multimodal Large Language Models (MLLMs) have achieved remarkable success across diverse vision-language tasks, yet their internal decision-making mechanisms remain insufficiently understood. Existing interpretability research has primarily focused on cross-modal attribution, identifying which image regions the model attends to during output generation. However, these approaches often overlook intra-modal dependencies. In the visual modality, attributing importance to isolated image patches ign…

"Come to work on time, know your lines and don't bump into the other actors."
—Spencer Tracy
#acting #coaching #inspiration

Carl Sagan’s Baloney Detection Kit: Tools for Thinking Critically & Knowing Pseudoscience When You See It
https://flip.it/iHmIqF

Reading the news and watching a real-world instance of “sends in a detachment of inexperienced riot police to the front lines of the protest” (Andor S02E08) play out in real time is *wild*
https://en.wikipedia.org/wiki/Who_Are_You?_(Andor)

Real software engineers don't like the idea of some inexplicable and
greasy hardware several aisles away that may stop working at any
moment. They have a great distrust of hardware people, and wish that
systems could be virtual at *___all* levels. They would like personal
computers (you know no one's going to trip over something and kill your
DFA in mid-transit), except that they need 8 megabytes to run their
Correctness Verification Aid packages.…

De #kieskompas scam is weer online.
Vul je alle vragen tegengesteld in aan de mening van het NSC, waar kom je dan terecht in het 'politieke landschap'? Jawel, vlak bij NSC (en bij CDA wat basically dezelfde partij is maar dan zonder al die minkukels).
Trap er niet in!

Evaluating the Limits of Large Language Models in Multilingual Legal Reasoning
Antreas Ioannou, Andreas Shiamishis, Nora Hollenstein, Nezihe Merve G\"urel
https://arxiv.org/abs/2509.22472

Evaluating the Limits of Large Language Models in Multilingual Legal Reasoning
In an era dominated by Large Language Models (LLMs), understanding their capabilities and limitations, especially in high-stakes fields like law, is crucial. While LLMs such as Meta's LLaMA, OpenAI's ChatGPT, Google's Gemini, DeepSeek, and other emerging models are increasingly integrated into legal workflows, their performance in multilingual, jurisdictionally diverse, and adversarial contexts remains insufficiently explored. This work evaluates LLaMA and Gemini on multilingual legal and non-l…

PRIME: Planning and Retrieval-Integrated Memory for Enhanced Reasoning
Hieu Tran, Zonghai Yao, Nguyen Luong Tran, Zhichao Yang, Feiyun Ouyang, Shuo Han, Razieh Rahimi, Hong Yu
https://arxiv.org/abs/2509.22315

PRIME: Planning and Retrieval-Integrated Memory for Enhanced Reasoning
Inspired by the dual-process theory of human cognition from \textit{Thinking, Fast and Slow}, we introduce \textbf{PRIME} (Planning and Retrieval-Integrated Memory for Enhanced Reasoning), a multi-agent reasoning framework that dynamically integrates \textbf{System 1} (fast, intuitive thinking) and \textbf{System 2} (slow, deliberate thinking). PRIME first employs a Quick Thinking Agent (System 1) to generate a rapid answer; if uncertainty is detected, it then triggers a structured System 2 rea…

Szabolcs-Szatmšr-Bereg's coat of arms is inspiring to me, i feel like it wouldnt take that many changes to make it something tuff enough for me to use for a project

Language Models Can Learn from Verbal Feedback Without Scalar Rewards
Renjie Luo, Zichen Liu, Xiangyan Liu, Chao Du, Min Lin, Wenhu Chen, Wei Lu, Tianyu Pang
https://arxiv.org/abs/2509.22638

Language Models Can Learn from Verbal Feedback Without Scalar Rewards
LLMs are often trained with RL from human or AI feedback, yet such methods typically compress nuanced feedback into scalar rewards, discarding much of their richness and inducing scale imbalance. We propose treating verbal feedback as a conditioning signal. Inspired by language priors in text-to-image generation, which enable novel outputs from unseen prompts, we introduce the feedback-conditional policy (FCP). FCP learns directly from response-feedback pairs, approximating the feedback-conditi…