Skip to content
TILens What matters today in tech v0.3.2
Theme

Daily edition · AI Research

The daily ledger

TILens turns technical updates into a focused daily brief: official releases, trusted reporting, and practitioner analysis, deduplicated and organized by topic.

18 Nov 2026 edition

329 articles · 6 sources · 325 papers ·

Top topics: AI · AI Research · Information Retrieval

AI Research AI Python

How AI Agents Become a Tool for Attackers, and How to Stop It

Managing Security When You’re Deploying AI, Part 1: Compromised credentials, salaries inferred from budgets, and doctored documents show why agents need permission checks before the model ever sees the data. The post…

Source: Towards Data Science Stephanie Kirmer
AI Research AI Python

Neurosymbolic Agents Explained, With a Z3 Example in Python

Starting from Air Canada's chatbot case, a tour of Logic-LM, Agent-C, and Amazon's checks shows the solver is sound but the English-to-logic step is not. The post Neurosymbolic Agents Explained, With a Z3 Example in…

Source: Towards Data Science Tomoharu Hayakawa
AI Research AI Python

How to Estimate Dose-Response Curves When DoubleML Falls Short

Simulations test DoubleML against Kennedy's doubly robust estimator, with partial pooling across 200 stores and a DiD fix for hidden confounders. The post How to Estimate Dose-Response Curves When DoubleML Falls Short…

Source: Towards Data Science Ryan OSullivan
AI Research AI

An Optimization Framework for Denoising Generation

arXiv:2611.22202v1 Announce Type: new Abstract: Noise-unconditioned denoising learns a single score, so generation can be viewed as gradient ascent on a fixed log-density landscape. This optimization interpretation…

Source: arXiv cs.LG Xinyu Zhou, Jiawei Zhang, Stephen J. Wright
AI Research AI

Learning Optimal Transport Plans Via Autoregressive Token Regression

arXiv:2611.22218v1 Announce Type: new Abstract: Optimal Transport (OT) provides a principled framework for finding the minimum cost coupling between probability distributions. Classical solvers remain the gold standard…

Source: arXiv cs.LG Ivan Jaen Marquez, Sotirios Panagiotis Chytas, Vikas Singh
AI Research AI

Token Commitment as a Control Surface for Steering Masked Diffusion Generation

arXiv:2611.21669v1 Announce Type: cross Abstract: Masked diffusion models repeatedly predict provisional tokens before a sampler decides which predictions become permanent. Existing inference-time steering methods…

Source: arXiv cs.LG Ambroise Aigueperse, Hugo Buurmeijer, Carmen Amo ALonso, Marco Pavone
AI Research AI

Margin Dynamics for Large Language Model Alignment

arXiv:2611.21766v1 Announce Type: new Abstract: Practitioners choose among LLM alignment methods largely by trial and error. We provide a theoretical foundation: each pairwise method defines a vector field on the…

Source: arXiv cs.LG Xingzi Xu, Mehmet Saygin Seyfioglu, Karim Bouyarmane
AI Research AI

From Constraints to Causality: Petri net Encodings for Explaining Binary Neural Network Inference

arXiv:2611.22396v1 Announce Type: cross Abstract: Binary Neural Networks (BNNs) are attractive for resource-constrained hardware and on-device components in autonomous systems. However, their discrete, non-linear…

Source: arXiv cs.LG Alex Chan (Newcastle University), Mohamed Tarraf (Newcastle University), Alex Yakovlev (Newcastle University), Rishad Shafik (Newcastle University)
AI Research AI

Rethinking "RL Generalizes, SFT Memorizes": The Role of SFT Data

arXiv:2611.14912v2 Announce Type: replace Abstract: Large language models trained with Reinforcement Learning (RL) with verifiable rewards exhibit strong reasoning ability and broad generalization, whereas models…

Source: arXiv cs.LG Yunlong Hou, Fengzhuo Zhang, Yuan Cheng, Jiachun Pan, Xingyao Li, Zhuoran Yang
AI Research AI

Exact Sequence Interpolation with Transformers

arXiv:2502.02270v4 Announce Type: replace Abstract: We prove that transformers can exactly interpolate datasets of finite input sequences in $\mathbb{R}^d$, $d\geq 2$, with corresponding output sequences of smaller or…

Source: arXiv cs.LG Albert Alcalde, Giovanni Fantuzzi, Enrique Zuazua

Showing 1 day · 329 items available