Adaptive Thinking: Large Language Models Know When to Think in Latent Space
Recent advances in large language models (LLMs) test-time computing have introduced the capability to perform intermediate chain-of-thought (CoT) reasoning (thinking) before generating answers. While increasing the thinking budget yields smooth performance improvements at inference time, the relatio...
DSO: Direct Steering Optimization for Bias Mitigation
Generative models are often deployed to make decisions on behalf of users, such as vision-language models (VLMs) identifying which person in a room is a doctor to help visually impaired individuals. Yet, VLM decisions are influenced by the perceived demographic attributes of people in the input, whi...
OpenAI Releases Privacy Filter: A 1.5B-Parameter Open-Source PII Redaction Model with 50M Active Parameters
OpenAI's Privacy Filter Is a 1.5B-Parameter PII Detector Built on a Distilled Decoder — And It Runs in Your Browser
The post OpenAI Releases Privacy Filter: A 1.5B-Parameter Open-Source PII Redaction Model with 50M Active Parameters appeared first on MarkTechPost.
NVIDIA Launches Nemotron 3 Nano Omni Model, Unifying Vision, Audio and Language for up to 9x More Efficient AI Agents
AI agent systems today juggle separate models for vision, speech and language — losing time and context as they pass data from one model to the other. Unveiled today, NVIDIA Nemotron 3 Nano Omni is an open multimodal model that brings these capabilities together into one system, enabling agents to d...
The Next Frontier of AI in Production Is Chaos Engineering
Blast-radius control tells you how much to break. Intent tells you what breaking it will teach. Only one of these has mature tooling.
The post The Next Frontier of AI in Production Is Chaos Engineering appeared first on Towards Data Science.
PyTorch NaNs Are Silent Killers — So I Built a 3ms Hook to Catch Them at the Exact Layer
NaNs don’t crash your training — they quietly destroy it.
After losing hours to a silent failure in a ResNet training run, I built a lightweight detector that pinpoints the exact layer and batch where things break. Using forward hooks and gradient checks, it catches issues early with minimal overhea...
The Spectral Lifecycle of Transformer Training: Transient Compression Waves, Persistent Spectral Gradients, and the Q/K--V Asymmetry
arXiv:2604.22778v1 Announce Type: new
Abstract: We present the first systematic study of weight matrix singular value spectra \emph{during} transformer pretraining, tracking full SVD decompositions of every weight matrix at 25-step intervals across three model scales (30M--285M parameters). We disc...
KARL: Mitigating Hallucinations in LLMs via Knowledge-Boundary-Aware Reinforcement Learning
arXiv:2604.22779v1 Announce Type: new
Abstract: Enabling large language models (LLMs) to appropriately abstain from answering questions beyond their knowledge is crucial for mitigating hallucinations. While existing reinforcement learning methods foster autonomous abstention, they often compromise ...
BiTA: Bidirectional Gated Recurrent Unit-Transformer Aggregator in a Temporal Graph Network Framework for Alert Prediction in Computer Networks
arXiv:2604.22781v1 Announce Type: new
Abstract: Proactive alert prediction in computer networks is critical for mitigating evolving cyber threats and enabling timely defensive actions. Temporal Graph Neural Networks (TGNs) provide a principled framework for modeling time-evolving interactions; howe...
arXiv:2604.22782v1 Announce Type: new
Abstract: Serving transformer language models with high throughput requires caching Key-Values (KVs) to avoid redundant computation during autoregressive generation. The memory footprint of KV caching is significant and heavily impacts serving costs. This work ...
Parameter Efficiency Is Not Memory Efficiency: Rethinking Fine-Tuning for On-Device LLM Adaptation
arXiv:2604.22783v1 Announce Type: new
Abstract: Parameter-Efficient Fine-Tuning (PEFT) has become the standard for adapting large language models (LLMs). In this work we challenge the wide-spread assumption that parameter efficiency equates memory efficiency and on-device adaptability. We show that...
An Intelligent Fault Diagnosis Method for General Aviation Aircraft Based on Multi-Fidelity Digital Twin and FMEA Knowledge Enhancement
arXiv:2604.22777v1 Announce Type: new
Abstract: Fault diagnosis of general aviation aircraft faces challenges including scarce real fault data, diverse fault types, and weak fault signatures. This paper proposes an intelligent fault diagnosis framework based on multi-fidelity digital twin, integrat...
PExA: Parallel Exploration Agent for Complex Text-to-SQL
arXiv:2604.22934v1 Announce Type: new
Abstract: LLM-based agents for text-to-SQL often struggle with latency-performance trade-off, where performance improvements come at the cost of latency or vice versa. We reformulate text-to-SQL generation within the lens of software test coverage where the ori...
The Power of Power Law: Asymmetry Enables Compositional Reasoning
arXiv:2604.22951v1 Announce Type: new
Abstract: Natural language data follows a power-law distribution, with most knowledge and skills appearing at very low frequency. While a common intuition suggests that reweighting or curating data towards a uniform distribution may help models better learn the...
On the Existence of an Inverse Solution for Preference-Based Reductions in Argumentation
arXiv:2604.22958v1 Announce Type: new
Abstract: Preference-based argumentation frameworks (PAFs) extend Dung's approach to abstract argumentation (AAFs) by encoding preferences over arguments. Such preferences control the transformation of attacks into defeats, and different approaches to doing so ...
Towards Causally Interpretable Wi-Fi CSI-Based Human Activity Recognition with Discrete Latent Compression and LTL Rule Extraction
arXiv:2604.22979v1 Announce Type: new
Abstract: We address Human Activity Recognition (HAR) utilizing Wi-Fi Channel State Information (CSI) under the joint requirements of causal interpretability, symbolic controllability, and direct operation on high-dimensional raw signals. Deep neural models ach...