Ranked Activation Shift for Post-Hoc Out-of-Distribution Detection
arXiv:2604.08572v1 Announce Type: new
Abstract: State-of-the-art post-hoc out-of-distribution detection methods rely on intermediate layer activation editing. However, they exhibit inconsistent performance across datasets and models. We show that this instability is driven by differences in the act...
OpenKedge: Governing Agentic Mutation with Execution-Bound Safety and Evidence Chains
arXiv:2604.08601v1 Announce Type: new
Abstract: The rise of autonomous AI agents exposes a fundamental flaw in API-centric architectures: probabilistic systems directly execute state mutations without sufficient context, coordination, or safety guarantees. We introduce OpenKedge, a protocol that re...
From Business Events to Auditable Decisions: Ontology-Governed Graph Simulation for Enterprise AI
arXiv:2604.08603v1 Announce Type: new
Abstract: Existing LLM-based agent systems share a common architectural failure: they answer from the unrestricted knowledge space without first simulating how active business scenarios reshape that space for the event at hand -- producing decisions that are fl...
RAMP: Hybrid DRL for Online Learning of Numeric Action Models
arXiv:2604.08685v1 Announce Type: new
Abstract: Automated planning algorithms require an action model specifying the preconditions and effects of each action, but obtaining such a model is often hard. Learning action models from observations is feasible, but existing algorithms for numeric domains ...
Parameterized Complexity Of Representing Models Of MSO Formulas
arXiv:2604.08707v1 Announce Type: new
Abstract: Monadic second order logic (MSO2) plays an important role in parameterized complexity due to the Courcelle's theorem. This theorem states that the problem of checking if a given graph has a property specified by a given MSO2 formula can be solved by a...
AI Weekly Issue #483: 100 years from now : The Ghost in the Contract
This is 100 Years From Now, a weekly series. Once a week, we skip ahead a century and imagine ordinary life in a world that's had a hundred years to absorb the things we're only beginning to build. No predictions — just honest speculation about where our choices lead.
This week: what happens when ac...
Cram Less to Fit More: Training Data Pruning Improves Memorization of Facts
This paper was accepted at the Workshop on Navigating and Addressing Data Problems for Foundation Models at ICLR 2026.
Large language models (LLMs) can struggle to memorize factual knowledge in their parameters, often leading to hallucinations and poor performance on knowledge-intensive tasks. In th...
Meta AI and KAUST Researchers Propose Neural Computers That Fold Computation, Memory, and I/O Into One Learned Model
Researchers from Meta AI and the King Abdullah University of Science and Technology (KAUST) have introduced Neural Computers (NCs) — a proposed machine form in which a neural network itself acts as the running computer, rather than as a layer sitting on top of one. The research team presents both a ...
A Coding Implementation of MolmoAct for Depth-Aware Spatial Reasoning, Visual Trajectory Tracing, and Robotic Action Prediction
In this tutorial, we walk through MolmoAct step by step and build a practical understanding of how action-reasoning models can reason in space from visual observations. We set up the environment, load the model, prepare multi-view image inputs, and explore how MolmoAct produces depth-aware reasoning...
Why storing and retrieving data isn’t enough to build reliable AI memory systems
The post Stop Treating AI Memory Like a Search Problem appeared first on Towards Data Science.
MiniMax Just Open Sourced MiniMax M2.7: A Self-Evolving Agent Model that Scores 56.22% on SWE-Pro and 57.0% on Terminal Bench 2
MiniMax has officially open-sourced MiniMax M2.7, making the model weights publicly available on Hugging Face. Originally announced on March 18, 2026, MiniMax M2.7 is the MiniMax’s most capable open-source model to date — and its first model to actively participate in its own development cycle, a me...
Four-Day Workweeks and Robot Taxes? OpenAI’s Radical Vision for the AI Future Is Turning Heads
It sounds like a late-night conversation with friends, what if artificial intelligence becomes so capable that we simply start working less. and what if we tax machines instead of people? It turns out you don’t need to dream big to make this conversation a reality. It’s now on the agenda of OpenAI. ...
The Robot Uprising Didn’t Happen. But Something Worse Did
More than 50,000 tech employees have lost their jobs so far this year. Asked why, most will say the same thing: artificial intelligence. Not because AI rose up and destroyed their workplaces, but because it assumed many of their responsibilities. And AI doesn’t draw a salary. So far this year, more ...
The End of Clicking? AI Is Quietly Turning Software Into Something That Just… Listens
Imagine never having to click through a software application again. Ever. The days of finding yourself staring blankly at the first screen of a new tool and wondering “What do I do now?” are disappearing. AI is quietly changing the way software is built and it’s a massive deal. Instead of software a...
Liquid AI Releases LFM2.5-VL-450M: a 450M-Parameter Vision-Language Model with Bounding Box Prediction, Multilingual Support, and Sub-250ms Edge Inference
Liquid AI just released LFM2.5-VL-450M, an updated version of its earlier LFM2-VL-450M vision-language model. The new release introduces bounding box prediction, improved instruction following, expanded multilingual understanding, and function calling support — all within a 450M-parameter footprint ...
Researchers from MIT, NVIDIA, and Zhejiang University Propose TriAttention: A KV Cache Compression Method That Matches Full Attention at 2.5× Higher Throughput
Long-chain reasoning is one of the most compute-intensive tasks in modern large language models. When a model like DeepSeek-R1 or Qwen3 works through a complex math problem, it can generate tens of thousands of tokens before arriving at an answer. Every one of those tokens must be stored in what is ...
Sam Altman responds to ‘incendiary’ New Yorker article after attack on his home
The OpenAI CEO's new blog post responds to both an apparent attack on his home and an in-depth New Yorker profile raising questions about his trustworthiness.
A deep-dive and practical guide to cross-encoders, advanced techniques, and why your retrieval pipeline deserves a second pass.
The post Advanced RAG Retrieval: Cross-Encoders & Reranking appeared first on Towards Data Science.
Agent harnesses are becoming the dominant way to build agents, and they are not going anywhere. These harnesses are intimately tied to agent memory. If you used a closed harness - especially if it’s behind a proprietary API - you are choosing to yield control of your agent&
Introduction to Reinforcement Learning Agents with the Unity Game Engine
A step-by-step interactive guide to one of the most vexing areas of machine learning.
The post Introduction to Reinforcement Learning Agents with the Unity Game Engine appeared first on Towards Data Science.
Understanding BERTopic: From Raw Text to Interpretable Topics
Topic modeling uncovers hidden themes in large document collections. Traditional methods like Latent Dirichlet Allocation rely on word frequency and treat text as bags of words, often missing deeper context and meaning. BERTopic takes a different route, combining transformer embeddings, clustering, ...
How Knowledge Distillation Compresses Ensemble Intelligence into a Single Deployable AI Model
Complex prediction problems often lead to ensembles because combining multiple models improves accuracy by reducing variance and capturing diverse patterns. However, these ensembles are impractical in production due to latency constraints and operational complexity. Instead of discarding them, Knowl...
Alibaba’s Tongyi Lab Releases VimRAG: a Multimodal RAG Framework that Uses a Memory Graph to Navigate Massive Visual Contexts
Retrieval-Augmented Generation (RAG) has become a standard technique for grounding large language models in external knowledge — but the moment you move beyond plain text and start mixing in images and videos, the whole approach starts to buckle. Visual data is token-heavy, semantically sparse relat...