4/10/2026
Privacy Attacks on Image AutoRegressive Models
Comprehensive privacy attack analysis on image autoregressive models, identifying membership inference and extraction vulnerabilities.
4/10/2026
$\texttt{SEM-CTRL}$: Semantically Controlled Decoding
Method for enforcing syntactic and semantic constraints in LLM decoding through MCTS-guided token-level control.
4/10/2026
OpenClassGen: A Large-Scale Corpus of Real-World Python Classes for LLM Research
Large-scale corpus of 324,843 Python classes from open-source projects for training and evaluating LLMs on code generation.
4/10/2026
ReCellTy: Domain-Specific Knowledge Graph Retrieval-Augmented LLMs Reasoning Workflow for Single-Cell Annotation
RAG-based LLM workflow using domain-specific knowledge graph for automated single-cell type annotation in biology.
4/10/2026
Are Sparse Autoencoders Useful for Java Function Bug Detection?
Study evaluating sparse autoencoders for detecting bugs in Java code, addressing software vulnerability detection.
4/10/2026
SealQA: Raising the Bar for Reasoning in Search-Augmented Language Models
Benchmark dataset (SealQA) for evaluating search-augmented LLMs on fact-seeking questions with conflicting or noisy search results.
4/10/2026
BacPrep: Lessons from Deploying an LLM-Based Bacalaureat Assessment Platform
Deployment case study of LLM-based platform for automated assessment of Romanian Bacalaureat exam questions using Gemini Flash.
4/10/2026
Video Parallel Scaling: Aggregating Diverse Frame Subsets for VideoLLMs
Inference method for VideoLLMs that processes multiple frame subsets in parallel to improve temporal detail without increasing context window.
4/10/2026
SeMoBridge: Semantic Modality Bridge for Efficient Few-Shot Adaptation of CLIP
Technique to improve CLIP few-shot classification by addressing modality gap through semantic bridging between image and text embeddings.
4/10/2026
Evaluating LLMs for Demographic-Targeted Social Bias Detection: A Comprehensive Benchmark Study
Benchmark for evaluating LLMs on detecting demographic-targeted social biases across diverse content types and demographics.
4/10/2026
Prompt reinforcing for long-term planning of large language models
Method to improve LLM performance in multi-turn conversations by reinforcing long-term planning and goal tracking through prompting.
4/10/2026
Mitigating Spurious Background Bias in Multimedia Recognition with Disentangled Concept Bottlenecks
Lightweight Disentangled Concept Bottleneck Model addressing bias in input-to-concept mapping for interpretable multimedia recognition.
4/10/2026
BADiff: Bandwidth Adaptive Diffusion Model
Framework enabling diffusion models to adapt generation quality based on real-time network bandwidth constraints in cloud-to-device scenarios.
4/10/2026
Rate-optimal Design for Anytime Best Arm Identification
Minimax optimal algorithm for best arm identification under fixed sampling budget with applications to A/B testing.
4/10/2026
Understanding Task Transfer in Vision-Language Models
Study of task transfer in Vision-Language Models examining how finetuning on one perception task affects performance on others.
4/10/2026
The Specification Trap: Why Static Value Alignment Alone Cannot Produce Robust Alignment
Philosophical analysis arguing static value alignment approaches cannot ensure robust AI alignment under capability scaling and distribution shift.
4/10/2026
OxEnsemble: Fair Ensembles for Low-Data Classification
OxEnsemble: Fair classification approach for low-data, imbalanced settings with demographic group constraints.
4/10/2026
Compact Example-Based Explanations for Language Models
Study on selecting minimal training data subsets for example-based explanations of language model predictions using influence estimation.
4/10/2026
Accordion-Thinking: Self-Regulated Step Summaries for Efficient and Readable LLM Reasoning
Accordion-Thinking: Framework enabling LLMs to self-regulate reasoning step granularity through dynamic summarization for efficient inference.
4/10/2026
Differentiable Logical Programming for Quantum Circuit Discovery and Optimization
Neuro-symbolic framework using differentiable logic programming to design and optimize quantum circuits.
4/10/2026
Stop Listening to Me! How Multi-turn Conversations Can Degrade LLM Diagnostic Reasoning
Evaluation of 17 LLMs showing diagnostic reasoning degrades across multi-turn conversations compared to single-turn benchmarks.
4/10/2026
HiCI: Hierarchical Construction-Integration for Long-Context Attention
HiCI: Hierarchical attention module for long-context language modeling, organizing information from local to global levels.
4/10/2026
CodecSight: Leveraging Video Codec Signals for Efficient Streaming VLM Inference
CodecSight optimizes streaming vision-language model inference by leveraging video codec signals for end-to-end efficiency.
4/10/2026
Distributional Open-Ended Evaluation of LLM Cultural Value Alignment Based on Value Codebook
DOVE benchmark for evaluating LLM cultural value alignment using open-ended generation, addressing limitations of multiple-choice formats.
4/10/2026
The Unreasonable Effectiveness of Data for Recommender Systems
Study investigating saturation points in recommender system performance as training dataset size increases, with reproducible Python implementation.
4/10/2026
The Detection-Extraction Gap: Models Know the Answer Before They Can Say It
Research: Chain-of-thought models generate 52-88% of tokens after answers are already recoverable, revealing inefficiency in reasoning.
4/10/2026
Some LLM routers are injecting malicious tool calls
Brief mention of LLM routers injecting malicious tool calls as security issue. Insufficient detail.
4/10/2026
Next SaaS replacement is an agent with a dashboard – kern
Analysis of AI agents as SaaS replacement with integrated database, logic, and UI.
4/10/2026
Show HN: Ctxlint – Lint your AGENTS.md for stale refs and token waste
Linter tool for AI agent context files and MCP configs. Detects stale references, token waste, hardcoded secrets. Cites research on context bloat reducing agent performance.
4/10/2026
Multica the open-source managed agents platform
Open-source platform for managing AI coding agents as teammates. Agents autonomously handle task assignment, code writing, and progress tracking without prompt copying.
4/10/2026
ActionParty: Multi-Subject Action Binding in Generative Video Games
Research on video generation models struggling with multi-subject action binding. Introduces ActionParty with per-subject state tokens to improve action accuracy.
4/10/2026
Google's AI Overviews spew false answers per hour, bombshell study reveals
Study reports Google's Gemini models generate inaccurate search results 9-15% of the time across 8,652 queries. Evaluates LLM reliability in production systems.
4/10/2026
Show HN: Facebook Marketplace MCP
Facebook Marketplace MCP integration allowing Claude to search and monitor deals via command line. Demonstrates AI agent tool use.
4/10/2026
How are you governing token spend and data exposure in prod AI agent loops?
Discussion question about managing token costs and data exposure in production AI agent systems. Operational concern without detailed analysis.
4/10/2026
Agentic Code Optimization via Compiler-LLM Cooperation
arXiv framework page for collaborative feature development. No technical content about the claimed compiler-LLM optimization research.
4/10/2026
Tend: Lightweight attention infrastructure for humans and AI agents
Tend: lightweight infrastructure for managing multiple concurrent AI agents and projects. Handles agent coordination, context switching, and multi-project workflows.
4/10/2026
5 LLM played Poker: Opus busted first, Grok won
Five LLM-based agents play Texas Hold'em with distinct personalities and reasoning styles.
4/10/2026
The role of LLMs in patch review
Discussion of LLM application to code patch review. Limited content, but relevant to developer tools and LLM applications.
4/10/2026
Stream your terminal. Watch AI agents work. Like Twitch, for shells
Tool for streaming terminal activity to observe AI agent execution. Visualization/monitoring for agent workflows.
4/10/2026
Show HN: SmolVM – open-source sandbox for coding and computer-use agents
Open-source local sandbox environment for AI agents on macOS/Linux. SDK/CLI enables persistent state, file I/O, pause/resume, and desktop environment interaction for agent workflows.
4/10/2026
Healthcare
AI resources for clinical workflows: evidence search, guideline reconciliation, and documentation support in healthcare.
4/10/2026
Financial services
Resources for evaluating, deploying, and scaling AI in regulated financial services environments.
4/10/2026
Applications of AI at OpenAI
Overview of OpenAI's evolution from research to consumer and developer products for AI applications.
4/9/2026
Trinity-Large-Thinking: Scaling an Open Source Frontier Agent
Trinity-Large-Thinking: open-source frontier reasoning model for long-horizon agents and tool calling. Apache 2.0 licensed weights on Hugging Face. Supports complex multi-turn interactions.
4/9/2026
Show HN: AnimTOON – AI that animates SVGs with 3-4x fewer tokens than OmniLottie
AnimTOON: Model for generating Lottie animations from text with 3-4x fewer tokens than alternatives, supporting skeletal animation.
4/9/2026
Bespoke OLAP: Using AI to Synthesize Workload-Specific DBMS Engines from Scratch
Research on using AI to synthesize workload-specific OLAP database engines optimized for specific query patterns.
4/9/2026
Show HN: Lingle – Voice agent to simulate zoom-based personal language lessons
Lingle is a voice-based AI agent that simulates personal language tutoring via conversation, addressing flexibility and pricing issues with traditional platforms.
4/9/2026
Show HN: Memoriki – LLM Wiki+MemPalace for persistent personal knowledge bases
Memoriki: Template combining LLM Wiki pattern with MemPalace MCP server for persistent personal knowledge bases with semantic search and temporal graphs.
4/9/2026
Show HN: Agent Tuning, using recursion to achieve predictable agent output
Agent Tuning technique using recursion to achieve predictable Claude Code agent output through self-observation and instruction editing.
4/9/2026
Open source died in March. It just doesn't know it yet
Analysis of open source supply chain attacks in March 2026 including credential theft and malicious library poisoning.