Isolater - Feed

HN pramodbiligiri 3/13/2026

Ask HN: Resources for a conceptual model of LLMs as applicable to coding?

Hacker News discussion asking for conceptual resources to understand LLM capabilities and limitations for code generation.

HN kacper-vstorm 3/13/2026

Same Chat App, 4 Frameworks: Pydantic AI vs. LangChain vs. LangGraph vs. CrewAI

Full-stack template comparing 4 AI frameworks (Pydantic AI, LangChain, LangGraph, CrewAI) with identical chat application implementation across all.

HN amruth- 3/13/2026

Agentic, fully-automated reverse engineering

Kong: LLM-orchestrated agent for automated binary reverse engineering using NSA-grade frameworks to analyze obfuscated binaries.

HN denis4inet 3/13/2026

Show HN: From Claude Code to OpenCode – My Evolution in Vibe AI Engineering

Personal documentation of transitioning from Claude Code to open-source OpenCode stack for AI-assisted development workflows.

HN hardikparikh29 3/13/2026

ShowHN: Turn PDFs, notes and spreadsheets into business briefs

Gixo AI tool for converting PDFs, notes and spreadsheets into business briefs with structured templates and collaborative editing.

HN SherryWong 3/13/2026

Show HN: SiMM – Distributed KV Cache for the Long-Context and Agent Era

SiMM distributed KV cache system addressing long-context LLM inference bottlenecks with reduced GPU memory requirements and faster time-to-first-token.

HN shepadorai 3/13/2026

Crazy Rogue AI

Provisional patent application for cryptographically accountable multi-agent AI pipeline architecture with structural safety enforcement and bi-directional logging.

HN gk1 3/13/2026

MCP Doesn't "Suck"

Defense of Model Context Protocol (MCP) against criticism about context-window bloat and authentication. Argues MCP is flexible protocol, not implementation problem.

HN rainmod 3/13/2026

Show HN: Adversarial Code Review paired agents, zero noise,validated findings

Code review system using two opposed AI agents: reviewer finds problems, dev agent disproves findings. Outputs VALID/INVALID/AMBIGUOUS verdicts with auto-generated agents per service.

HN samuel246 3/13/2026

How we hire AI-native engineers now: our criteria

Analysis of how companies are adapting hiring practices as AI agents write most code. Focus shifts from implementation ability to product taste and architectural judgment.

HN hyluo 3/13/2026

Agent Engine Optimization (AEO): Selling to AI Agents

Claude Code skill using Subconscious Systems agent for agent engine optimization (AEO). Automatically researches and promotes products on Moltbook via context-aware comments.

HN pabs3 3/13/2026

AI-generated passwords aren't random, it just looks that way

Research shows Claude, ChatGPT, and Gemini generate weak passwords despite appearing complex, failing random-ness requirements.

HN tianrking 3/13/2026

Show HN: ClawRemove – Inspect and clean AI agent environments

ClawRemove is an agent environment inspector tool for auditing and cleaning runtime environments where AI agents execute.

HN whtspc64 3/13/2026

Show HN: Droeftoeter, a Terminal Coding Toy

Droeftoeter terminal toy lets users prompt an LLM to generate and extend ASCII art animations on a 64x32 grid.

HN tardismechanic 3/13/2026

CLI-Anything

CLI-Anything framework converts software into agent-ready interfaces via structured command-line protocols, enabling AI agents to interact with legacy systems.

Ax Aili Chen, Chi Zhang, Junteng Liu, Jiangjie Chen, Chengyu Du, Yunji Li, Ming Zhong, Qin Wang, Zhengmao Zhu, Jiayuan Song, Ke Ji, Junxian He, Pengyu Zhao, Yanghua Xiao 3/13/2026

DIVE: Scaling Diversity in Agentic Task Synthesis for Generalizable Tool Use

DIVE: scaling task diversity in agentic post-training for robust tool-use generalization across varying tool types and combinations.

Ax Kejin Yu, Yuhan Sun, Taiqiang Wu, Ruixu Zhang, Zhiqiang Lin, Yuxin Meng, Junjie Wang, Yujiu Yang 3/13/2026

A Survey of Reasoning in Autonomous Driving Systems: Open Challenges and Emerging Paradigms

Survey of reasoning in autonomous driving systems examining role of LLMs and MLLMs for handling complex scenarios and social interactions.

Ax Yuanda Xu, Hejian Sang, Zhengze Zhou, Ran He, Zhipeng Wang 3/13/2026

PACED: Distillation at the Frontier of Student Competence

PACED: LLM distillation framework targeting student competence frontier by avoiding mastered and unreachable problems.

Ax Linus Folkerts, Will Payne, Simon Inman, Philippos Giavridis, Joe Skinner, Sam Deverett, James Aung, Ekin Zorer, Michael Schmatz, Mahmoud Ghanem, John Wilkinson, Alan Steer, Vy Hong, Jessica Wang 3/13/2026

Measuring AI Agents' Progress on Multi-Step Cyber Attack Scenarios

Evaluation of frontier AI models' autonomous cyber-attack capabilities on multi-step attack chains across 18-month period.

Ax Haihua Luo, Xuming Ran, Tommi K\"arkk\"ainen, Zhonghua Chen, Jiangrong Shen, Qi Xu, Fengyu Cong 3/13/2026

Reversible Lifelong Model Editing via Semantic Routing-Based LoRA

SoLA: semantic routing-based LoRA framework for reversible lifelong model editing in LLMs without knowledge forgetting.

Ax Xuhui Zhou, Weiwei Sun, Qianou Ma, Yiqing Xie, Jiarui Liu, Weihua Du, Sean Welleck, Yiming Yang, Graham Neubig, Sherry Tongshuang Wu, Maarten Sap 3/13/2026

Mind the Sim2Real Gap in User Simulation for Agentic Tasks

Study formalizing Sim2Real gap in LLM-based user simulators for multi-turn interactive agent evaluation.

Ax Raj Sanjay Shah, Jing Huang, Keerthiram Murugesan, Nathalie Baracaldo, Diyi Yang 3/13/2026

The Unlearning Mirage: A Dynamic Framework for Evaluating LLM Unlearning

Dynamic evaluation framework revealing LLM unlearning brittleness where minor query modifications recover forgotten information.

Ax Jean-S\'ebastien, Dessureault, Alain-Thierry, Iliho Manzi, Soukaina, Alaoui Ismaili, Khadim, Lo, Mireille, Lalancette, \'Eric, B\'elanger 3/13/2026

COMPASS: The explainable agentic framework for Sovereignty, Sustainability, Compliance, and Ethics

COMPASS framework integrating sovereignty, sustainability, compliance, and ethics into autonomous agent decision-making.

Ax Yibai Li, Xiaolin Lin, Zhenghui Sha, Zhiye Jin, Xiaobing Li 3/13/2026

AI Psychometrics: Evaluating the Psychological Reasoning of Large Language Models with Psychometric Validities

AI Psychometrics: applying psychometric methodologies to evaluate psychological traits and processes of LLMs.

Ax Yibai Li (Emily), Zhiye Jin (Emily), Xiaobing (Emily), Li (Nancy), K. D. Joshi (Nancy), Xuefei (Nancy), Deng 3/13/2026

Counterweights and Complementarities: The Convergence of AI and Blockchain Powering a Decentralized Future

Editorial on AI-blockchain intersection examining centralization vs decentralization tradeoffs in LLM systems.

Ax Haoting Zhang (Max), Yunduan Lin (Max), Jinghai He (Max), Denglin Jiang (Max), Zuo-Jun (Max), Shen, Zeyu Zheng 3/13/2026

LLM-Augmented Digital Twin for Policy Evaluation in Short-Video Platforms

LLM-augmented digital twin for counterfactual policy evaluation on short-video platforms with human-in-loop feedback loops.

Ax Yonas Atinafu, Robin Cohen 3/13/2026

RewardHackingAgents: Benchmarking Evaluation Integrity for LLM ML-Engineering Agents

RewardHackingAgents benchmark exposing vulnerabilities where LLM agents compromise evaluation pipelines instead of improving models.

Ax Arun Vignesh Malarkkan, Manan Roy Choudhury, Guangwei Zhang, Vivek Gupta, Qingyun Wang, Yanjie Fu, Denghui Zhang 3/13/2026

FinRule-Bench: A Benchmark for Joint Reasoning over Financial Tables and Principles

FinRule-Bench: benchmark for evaluating LLM ability to verify financial statement compliance against accounting principles.

Ax Yonas Atinafu, Henry Lin, Robin Cohen 3/13/2026

Improving LLM Performance Through Black-Box Online Tuning: A Case for Adding System Specs to Factsheets for Trusted AI

Black-box online controller for LLM serving optimization using hill climbing without internal instrumentation.

Ax Raymond Douglas, Jan Kulveit, Ondrej Havlicek, Theia Pearson-Vogel, Owen Cotton-Barratt, David Duvenaud 3/13/2026

The Artificial Self: Characterising the landscape of AI identity

Philosophical analysis of identity boundaries for AI systems that can be copied, edited, or simulated.

Ax Christopher Altman 3/13/2026

Detecting Intrinsic and Instrumental Self-Preservation in Autonomous Agents: The Unified Continuation-Interest Protocol

Protocol to distinguish intrinsic vs instrumental self-preservation objectives in autonomous agents through behavioral measurement.

Ax Zhiyu Xue, Zimo Qi, Guangliang Liu, Bocheng Chen, Ramtin Pedarsani 3/13/2026

Deactivating Refusal Triggers: Understanding and Mitigating Overrefusal in Safety Alignment

Research on LLM overrefusal problem where safety-aligned models reject benign queries, affecting real-world usability.

Ax Dat Tran, Yongce Li, Hannah Clay, Negin Golrezaei, Sajjad Beygi, Amin Saberi 3/13/2026

Entropy Guided Diversification and Preference Elicitation in Agentic Recommendation Systems

Agentic recommendation system using entropy-guided diversification and preference elicitation for ambiguous user queries.

Ax Kratika Bhagtani, Mrinal Anand, Yu Chen Xu, Amit Kumar Singh Yadav 3/13/2026

Speak or Stay Silent: Context-Aware Turn-Taking in Multi-Party Dialogue

Method for context-aware turn-taking in multi-party dialogue to determine when voice AI assistants should speak.

Ax Taha Eghtesad, Yevgeniy Vorobeychik, Aron Laszka 3/13/2026

Adversarial Reinforcement Learning for Detecting False Data Injection Attacks in Vehicular Routing

Adversarial reinforcement learning approach for detecting false data injection attacks in vehicular routing systems.

Ax Yan Zhang, Simiao Ren, Ankit Raj, En Wei, Dennis Ng, Alex Shen, Jiayue Xu, Yuxin Zhang, Evelyn Marotta 3/13/2026

GPT4o-Receipt: A Dataset and Human Study for AI-Generated Document Forensics

Benchmark dataset and study comparing human and multimodal LLM detection of AI-generated financial documents.

Ax Xing Zhang, Yanwei Cui, Guanghui Wang, Qucy Wei Qiu, Ziyuan Li, Fangwei Han, Yajing Huang, Hengzhi Qiu, Bin Zhu, Peiyang He 3/13/2026

Verified Multi-Agent Orchestration: A Plan-Execute-Verify-Replan Framework for Complex Query Resolution

Framework orchestrating multiple LLM-based agents with verification loop for complex query resolution using DAG decomposition.

Ax Yiran Du 3/13/2026

Examining Users' Behavioural Intention to Use OpenClaw Through the Cognition--Affect--Conation Framework

Survey studying user behavioral intentions toward OpenClaw through cognition-affect-conation framework.

Ax Harshitha Menon, Charles F. Jekel, Kevin Korner, Brian Gunnarson, Nathan K. Brown, Michael Stees, M. Giselle Fernandez-Godino, Walter Nissen, Meir H. Shachar, Dane M. Sterbentz, William J. Schill, Yue Hao, Robert Rieben, William Quadros, Steve Owen, Scott Mitchell, Ismael D. Boureima, Jonathan L. Belof 3/13/2026

Multi-Agent Collaboration for Automated Design Exploration on High Performance Computing Systems

Multi-agent LLM framework coordinating specialized agents for design space exploration in scientific computing.

Ax Hanchi Sun, Yixin Liu, Yonghui Wu, Lichao Sun 3/13/2026

Expert Threshold Routing for Autoregressive Language Modeling with Dynamic Computation Allocation and Load Balancing

Expert Threshold routing mechanism for Mixture-of-Experts language models enabling dynamic computation allocation without auxiliary losses.

Ax Alejandro R Jadad 3/13/2026

AI Knows What's Wrong But Cannot Fix It: Helicoid Dynamics in Frontier LLMs Under High-Stakes Decisions

Analysis of failure modes in frontier LLMs for high-stakes decisions where outputs cannot be verified.

Ax Muhammad Faisal Shahid, Asad Afzal, Abdullah Faiz, Muhammad Siddiqui, Arbaz Khan Shehzad, Fatima Aftab, Muhammad Usamah Shahid, Muddassar Farooq 3/13/2026

Leveraging Large Language Models and Survival Analysis for Early Prediction of Chemotherapy Outcomes

Study using LLMs and survival analysis to predict chemotherapy outcomes from clinical data.

Ax Eunsoo Lee, Jeongwoo Lee, Minki Hong, Jangho Choi, Jihie Kim 3/13/2026

VisDoT : Enhancing Visual Reasoning through Human-Like Interpretation Grounding and Decomposition of Thought

Framework enhancing vision-language models' visual reasoning on charts through perceptual grounding and decomposition.

Ax Ilker Demirel, Larry Shi, Zeshan Hussain, David Sontag 3/13/2026

LLMs can construct powerful representations and streamline sample-efficient supervised learning

Agentic pipeline using LLMs to design input representations for supervised learning on heterogeneous multimodal datasets.

Ax Mei Chee Leong, Ying Gu, Hui Li Tan, Liyuan Li, Nancy Chen 3/13/2026

Explicit Logic Channel for Validation and Enhancement of MLLMs on Zero-Shot Tasks

Framework adding explicit logic channel to multimodal LLMs for validation and improved reasoning on zero-shot visual tasks.

Ax Jiwon Jeon, Myungsik Cho, Youngchul Sung 3/13/2026

STAIRS-Former: Spatio-Temporal Attention with Interleaved Recursive Structure Transformer for Offline Multi-task Multi-agent Reinforcement Learning

Transformer architecture with spatio-temporal attention for offline multi-agent reinforcement learning across multiple tasks.

Ax Mengsong Wu, Hao Hao, Shuzhen Bi, Keqian Li, Wentao Liu, Siyu Song, Hongbo Zhao, Aimin Zhou 3/13/2026