3/12/2026
Show HN: Restailor – open-source AI job fit/resume tailor/job tracker
Open-source resume and job application tool using LLM integrations from multiple providers for tailoring and job tracking.
3/12/2026
Which Model Uncensoring Method Works Best?
Comparison of three methods (Ablation, Heretic, Obliteratus) for removing refusal behaviors from LLMs.
3/12/2026
Show HN: AutoICD API – AI clinical coding platform for ICD-10 and SNOMED
AI platform for converting clinical medical text to ICD-10 and SNOMED-CT codes using multi-layer ML architecture with custom models.
3/12/2026
Agentic Control Center for Data Product Optimization
System automating data product improvement using specialized AI agents to generate supporting assets and example queries without manual expertise.
3/12/2026
Hybrid Self-evolving Structured Memory for GUI Agents
Structured memory system for GUI agents combining trajectory-based and learned memory to handle long-horizon workflows and interface diversity.
3/12/2026
HEAL: Hindsight Entropy-Assisted Learning for Reasoning Distillation
HEAL method for distilling reasoning from large reasoning models into smaller models using hindsight entropy to overcome teacher ceiling limitations.
3/12/2026
Beyond Scalars: Evaluating and Understanding LLM Reasoning via Geometric Progress and Stability
TRACED framework evaluates LLM reasoning reliability through geometric kinematics measuring progress and stability rather than scalar probabilities.
3/12/2026
Verbalizing LLM's Higher-order Uncertainty via Imprecise Probabilities
Framework using imprecise probabilities to capture and verbalize higher-order uncertainty in LLMs beyond classical probabilistic methods.
3/12/2026
Resource-constrained Amazons chess decision framework integrating large language models and graph attention
Lightweight hybrid framework combining LLMs and graph attention networks for resource-constrained game-playing decision-making.
3/12/2026
IH-Challenge: A Training Dataset to Improve Instruction Hierarchy on Frontier LLMs
Training dataset addressing instruction hierarchy in frontier LLMs to defend against jailbreaks, prompt injections, and agentic attacks.
3/12/2026
Adaptive RAN Slicing Control via Reward-Free Self-Finetuning Agents
Approach using reward-free self-finetuning agents for adaptive RAN slicing control in network systems overcoming context window limitations.
3/12/2026
CUAAudit: Meta-Evaluation of Vision-Language Models as Auditors of Autonomous Computer-Use Agents
Meta-evaluation framework using vision-language models to audit and evaluate autonomous computer-use agents at scale beyond static benchmarks.
3/12/2026
Does LLM Alignment Really Need Diversity? An Empirical Study of Adapting RLVR Methods for Moral Reasoning
Empirical study examining whether LLM alignment for moral reasoning requires diversity-seeking algorithms versus reward-maximizing policies using RLVR methods.
3/12/2026
Trajectory-Informed Memory Generation for Self-Improving Agent Systems
Framework for extracting actionable learnings from LLM agent execution trajectories to improve future performance and reduce repeated errors.
3/12/2026
FAME: Formal Abstract Minimal Explanation for Neural Networks
FAME proposes formal abstract minimal explanations for neural networks using abstract interpretation, scaling to large models with reduced explanation size.
3/12/2026
Emulating Clinician Cognition via Self-Evolving Deep Clinical Research
DxEvolve is a self-evolving diagnostic agent mimicking clinician cognition through interactive deep clinical research with auditable improvement mechanisms.
3/12/2026
Nurture-First Agent Development: Building Domain-Expert AI Agents Through Conversational Knowledge Crystallization
Framework for building domain-expert AI agents through conversational knowledge crystallization, shifting from code-first/prompt-first approaches.
3/12/2026
A Hybrid Knowledge-Grounded Framework for Safety and Traceability in Prescription Verification
PharmGraph-Auditor system for safe medication verification using LLMs with knowledge graphs for traceability and complex reasoning in healthcare.
3/12/2026
Decoupling Reasoning and Confidence: Resurrecting Calibration in Reinforcement Learning from Verifiable Rewards
Research addressing calibration degeneration in reinforcement learning from verifiable rewards by decoupling reasoning and confidence in LLM training.
3/12/2026
One Model, Many Skills: Parameter-Efficient Fine-Tuning for Multitask Code Analysis
Parameter-efficient fine-tuning approach enabling single LLM to handle multiple code analysis tasks with reduced computational cost.
3/12/2026
Explainable LLM Unlearning Through Reasoning
Research on LLM unlearning through reasoning to remove undesirable knowledge while mitigating safety, copyright, and privacy issues.
3/12/2026
AraModernBERT: Transtokenized Initialization and Long-Context Encoder Modeling for Arabic
AraModernBERT adapts ModernBERT architecture to Arabic with transtokenized embedding initialization and native 8,192 token context support.
3/12/2026
MoE-SpAc: Efficient MoE Inference Based on Speculative Activation Utility in Heterogeneous Edge Scenarios
Research on efficient MoE model inference on edge devices using speculative activation utility and lookahead sensors for memory management.
3/12/2026
The Dunning-Kruger Effect in Large Language Models: An Empirical Study of Confidence Calibration
Empirical study investigating whether LLMs exhibit Dunning-Kruger effect patterns, analyzing confidence calibration across state-of-the-art models.
3/12/2026
Quantifying Hallucinations in Language Language Models on Medical Textbooks
Research paper quantifying hallucination rates in LLMs on medical textbook QA tasks, measuring factually incorrect claims against fixed evidence sources.
3/12/2026
Evolving Demonstration Optimization for Chain-of-Thought Feature Transformation
Demonstration optimization method using evolutionary algorithms for chain-of-thought feature transformation in data-centric AI tasks.
3/12/2026
Causally Grounded Mechanistic Interpretability for LLMs with Faithful Natural-Language Explanations
Pipeline for mechanistic interpretability combining activation patching with natural language explanations of LLM internal circuits.
3/12/2026
The System Hallucination Scale (SHS): A Minimal yet Effective Human-Centered Instrument for Evaluating Hallucination-Related Behavior in Large Language Models
System Hallucination Scale (SHS): human-centered measurement instrument for evaluating factual unreliability and hallucination behavior in LLMs.
3/12/2026
A Two-Stage Architecture for NDA Analysis: LLM-based Segmentation and Transformer-based Clause Classification
Two-stage NDA analysis system using LLaMA for clause extraction and transformers for contract clause classification.
3/12/2026
TAMUSA-Chat: A Domain-Adapted Large Language Model Conversational System for Research and Responsible Deployment
Framework for building domain-adapted LLM conversational systems using fine-tuning, RAG, and evaluation methodologies for institutional deployment.
3/12/2026
CEI: A Benchmark for Evaluating Pragmatic Reasoning in Language Models
CEI benchmark with 300 validated scenarios for evaluating pragmatic reasoning and contextual inference in LLMs.
3/12/2026
Evaluating Adjective-Noun Compositionality in LLMs: Functional vs Representational Perspectives
Analysis of adjective-noun compositionality in LLMs using prompt-based and representational methods revealing performance-internal state divergence.
3/12/2026
Context Over Compute Human-in-the-Loop Outperforms Iterative Chain-of-Thought Prompting in Interview Answer Quality
Study comparing human-in-the-loop approaches with chain-of-thought prompting for behavioral interview evaluation using LLMs.
3/12/2026
There Are No Silly Questions: Evaluation of Offline LLM Capabilities from a Turkish Perspective
Evaluation of offline LLM capabilities for Turkish heritage language education focusing on pedagogical safety and robustness.
3/12/2026
Empathy Is Not What Changed: Clinical Assessment of Psychological Safety Across GPT Model Generations
Clinical evaluation of GPT model generations for psychological safety across emotionally challenging conversational scenarios.
3/12/2026
Automated evaluation of LLMs for effective machine translation of Mandarin Chinese to English
Automated evaluation framework for assessing LLM machine translation quality from Mandarin Chinese to English.
3/12/2026
A Retrieval-Augmented Language Assistant for Unmanned Aircraft Safety Assessment and Regulatory Compliance
Retrieval-augmented assistant for unmanned aircraft safety assessment and regulatory compliance automation.
3/12/2026
Leveraging Wikidata for Geographically Informed Sociocultural Bias Dataset Creation: Application to Latin America
Dataset creation method using Wikidata to detect sociocultural bias in LLMs for Latin American languages and contexts.
3/12/2026
SpreadsheetArena: Decomposing Preference in LLM Generation of Spreadsheet Workbooks
Benchmark platform for evaluating LLM performance on end-to-end spreadsheet generation with explicit and implicit user constraints.
3/12/2026
GATech at AbjadMed: Bidirectional Encoders vs. Causal Decoders: Insights from 82-Class Arabic Medical Classification
Arabic medical text classification system using fine-tuned AraBERTv2 encoder benchmarked against multilingual and Arabic-specific models.
3/12/2026
Personalized Group Relative Policy Optimization for Heterogenous Preference Alignment
Method for personalizing LLM alignment to diverse user preferences beyond single global objectives using group-relative policy optimization.
3/12/2026
FERRET: Framework for Expansion Reliant Red Teaming
Automated red teaming framework for generating multi-modal adversarial conversations to test LLM robustness with expansion techniques.
3/12/2026
Measuring and Eliminating Refusals in Military Large Language Models
Research on measuring and reducing safety refusals in military LLMs to enable accurate information provision in combat scenarios.
3/12/2026
Assessing Cognitive Biases in LLMs for Judicial Decision Support: Virtuous Victim and Halo Effects
Study assessing cognitive biases in LLMs (virtuous victim effect, halo effect) relevant to judicial decision support applications.
3/12/2026
DeliberationBench: A Normative Benchmark for the Influence of Large Language Models on Users' Views
DeliberationBench benchmark for measuring persuasive influence of LLMs on user beliefs using deliberative polling methodology.
3/12/2026
Defining AI Models and AI Systems: A Framework to Resolve the Boundary Problem
Policy analysis defining 'AI model' and 'AI system' through systematic review of 896 papers and 80+ regulatory/technical documents.
3/12/2026
RedFuser: An Automatic Operator Fusion Framework for Cascaded Reductions on AI Accelerators
RedFuser automatic operator fusion framework for cascaded reduction operations in attention mechanisms on AI accelerators.
3/12/2026
How to Count AIs: Individuation and Liability for AI Agents
Policy paper on legal and accountability frameworks for identifying and governing autonomous AI agents in economic systems.
3/12/2026
The DMA Streaming Framework: Kernel-Level Buffer Orchestration for High-Performance AI Data Paths
Linux kernel module dmaplane for buffer orchestration and lifecycle management in high-performance AI data paths.
3/12/2026
Architecture-Aware LLM Inference Optimization on AMD Instinct GPUs: A Comprehensive Benchmark and Deployment Study
Benchmark study of LLM inference optimization on AMD Instinct GPUs for 235B-1T parameter models with architecture-specific deployment guidance.