The paper introduces LLM-Microscope, a toolkit for measuring how Large Language Models encode contextual information at the token level. It shows that seemingly minor tokens—such as determiners, stopwords, and punctuation—carry surprisingly high contextual weight, and removing them degrades performance on benchmarks like MMLU and BABILong-4k. The study also finds a strong link between contextualization and linearity, indicating that the transformation between layers can be approximated by a single linear mapping when tokens are well contextualized.
By Anton Razzhigaev, Matvey Mikhalchuk, Temurbek Rahmatullaev, Elizaveta Goncharova, Polina Druzhinina, Ivan Oseledets, Andrey Kuznetsov
Why use it? How to implement it?
By Shuai Guo
arXiv:2511. 21594v3 Announce Type: replace Abstract: Large language models (LLMs) achieve state-of-the-art results across many natural language tasks, but their internal mechanisms remain difficult to interpret.
By Alex Ning, Vainateya Rangaraju, Yen-Ling Kuo
The article "How to Fine-Tune an LLM: An End-to-End Guide" offers a practical, hands‑on walkthrough for fine‑tuning large language models in real‑world scenarios. It covers the entire process from data preparation to deployment, providing readers with actionable steps to adapt LLMs to specific tasks. The guide is aimed at practitioners looking to implement fine‑tuning in a structured, end‑to‑end manner.
By Sam Black
Enterprise Document Intelligence [Vol. 1 #7C] - One LLM call ranks the candidates with reasons.
By angela shi
arXiv:2509. 25359v2 Announce Type: replace-cross Abstract: We present a systematic stress-test of geometric metrics for LLM evaluation.
By Viacheslav Yusupov, Anna Antipina, Ameliia Alaeva, Danil Maksimov, Anna Vasileva, Tatyana Zaitseva, Alina Ermilova, Evgeny Burnaev, Egor Shvetsov
Increase productivity with your LLMs The post How to Effectively Align with Claude Code appeared first on Towards Data Science .
By Eivind Kjosbakken
The article explains how transformers, which rely on self‑attention, can lose the natural order of time‑series data when fed scalar observations. It discusses the role of positional encoding in re‑introducing sequence order and provides a visual guide to illustrate this concept.
By Gurjinder Kaur
t0-alpha is a decoder-style patch transformer for probabilistic time-series forecasting. Raw series are split into 32-step patches, embedded, processed through causal time-attention and group-attention layers, and decoded into future quantiles rather than a single point forecast.
By Sean Moran
The paper studies how transformer representations evolve across layers by examining the intrinsic dimensionality (ID) of token embeddings and their neighborhood structures. It finds that closed‑class tokens expand and collapse earlier than open‑class tokens, and that these changes are linked to shifts in local geometry. The authors compare encoder and decoder models, showing distinct layer‑wise behaviors, and demonstrate that geometric features alone can predict a token’s part‑of‑speech and reveal how semantic content changes across layers.
By Samuele Vallisa, Federico Ravenda, Claudio Palominos, Rui He, Andrea Raballo, Antonietta Mira, Philipp Homan, Wolfram Hinzen
arXiv:2608. 01283v1 Announce Type: new Abstract: All Transformer-based large language models compute attention via the Euclidean inner product, an architectural choice that Dong et al.
By Sen Song
Getting reliable, readable responses out of your LLM, and knowing which tool to reach for The post Structured Outputs with LLMs: JSON Mode, Function Calling, and When to Use Each appeared first on Towards Data Science .
By Maria Mouschoutzi