6/15/2026
Show HN: Applora – extracting product feedback from Shopify app reviews
Applora tool extracting product feedback and merchant pain points from negative Shopify app reviews using AI analysis.
6/15/2026
Show HN: AgentBridge – translate and govern calls between AI agent protocols
AgentBridge protocol mesh for translating and governing calls between AI agent protocols with audit trails and budget controls.
6/15/2026
AI Agents Don't Need SMS APIs. They Need Infrastructure
Article discussing infrastructure requirements for AI agents beyond simple API integrations.
6/15/2026
Fugee, an agentic AI assistant for displaced people and asylum seekers [video]
Fugee agentic AI assistant designed to help displaced people and asylum seekers navigate systems.
6/15/2026
Deep Dense Exploration for LLM Reinforcement Learning via Pivot-Driven Resampling
Deep Dense Exploration method improves reinforcement learning for LLMs by using pivot-driven resampling to discover high-quality trajectories more efficiently than existing tree-based approaches.
6/15/2026
FedRot-LoRA: Mitigating Rotational Misalignment in Federated LoRA
FedRot-LoRA addresses rotational misalignment in federated fine-tuning of LLMs, proposing solutions to improve aggregation accuracy and training stability in decentralized settings.
6/15/2026
Generalized Discrete Diffusion with Self-Correction
Pretraining-based self-correction for discrete diffusion models via multi-step uniform-absorbing objective.
6/15/2026
IntSeqBERT: Learning Arithmetic Structure in OEIS via Modulo-Spectrum Embeddings
Transformer encoder for integer sequence modeling using modulo-spectrum embeddings to handle arithmetic structure in OEIS.
6/15/2026
The Curse and Blessing of Mean Bias in FP4-Quantized LLM Training
Analysis of mean bias effects in FP4 quantized LLM training, identifying rank-one activation outliers as training fragility source.
6/15/2026
Temporal Straightening for Latent Planning
Representation learning approach for latent planning using temporal straightening inspired by perceptual processing.
6/15/2026
TabKD: Tabular Knowledge Distillation through Interaction Diversity of Learned Feature Bins
Data-free knowledge distillation method for tabular models leveraging learned feature bin interaction diversity.
6/15/2026
AcceRL: A Distributed Asynchronous Reinforcement Learning and World Model Framework for Vision-Language-Action Models
Distributed asynchronous reinforcement learning framework for Vision-Language-Action models eliminating synchronization bottlenecks.
6/15/2026
Dynamic Free-Rider Detection in Federated Learning via Simulated Attack Patterns
Detection method for free-riders in federated learning using simulated attack patterns and parameter evolution analysis.
6/15/2026
STaR-DRO: Stateful Tsallis Reweighting for Group-Robust Structured Prediction
Framework for ontology-constrained LLM generation with group robustness and label imbalance handling via reweighting.
6/15/2026
Can LLMs Accurately Score Medical Diagnoses and Clinical Reasoning?
Evaluation of LLM panels as adjudicators for scoring medical diagnoses and clinical reasoning on real hospital cases.
6/15/2026
LEPO: Latent Reasoning Policy Optimization for Large Language Models
Latent reasoning optimization method for LLMs using Gumbel-Softmax to enable diverse reasoning paths and exploration.
6/15/2026
Knowing When to Quit: A Principled Framework for Dynamic Abstention in LLM Reasoning
Framework for dynamic abstention in LLM reasoning, enabling mid-generation termination of unpromising reasoning traces.
6/15/2026
Sub-Token Routing for KV Cache Compression
KV cache compression technique using sub-token routing for efficient transformer inference in long-context and multimodal generation.
6/15/2026
Scalable Production Scheduling: Linear Complexity via Unified Homogeneous Graphs
Graph-based reinforcement learning approach for job shop scheduling with linear complexity and improved scalability.
6/15/2026
OGPO: Sample Efficient Full-Finetuning of Generative Control Policies
Sample-efficient algorithm for fine-tuning generative control policies in robot learning using off-policy critics.
6/15/2026
Quantile-Free Uncertainty Quantification in Graph Neural Networks
Uncertainty quantification method for graph neural networks without quantile estimation, using efficient prediction intervals.
6/15/2026
Towards Steering without Sacrifice: Principled Training of Steering Vectors for Prompt-only Interventions
Framework for training steering vectors to control LLM behavior without sacrificing generation quality or requiring per-vector tuning.
6/15/2026
Where's the Plan? Locating Latent Planning in Language Models with Lightweight Mechanistic Interventions
Mechanistic study of planning in large language models using linear probing and activation patching across multiple model scales.
6/15/2026
A Composite Activation Function for Learning Stable Binary Representations
Novel activation function enabling stable training of binary neural networks for improved computational efficiency and interpretability.
6/15/2026
EmoMind: Decoding Affective Captions from Human Brain fMRI
Brain-to-text system using fMRI to decode affective captions with emotional content rather than just semantic information.
6/15/2026
Learning Variable-Length Tokenization for Generative Recommendation
Proposes variable-length tokenization for generative recommendation systems, discovering that popular items need different encoding capacity.
6/15/2026
Exact Linear Attention
Exact Linear Attention achieves linear complexity Transformer attention via kernel decomposition without approximation error.
6/15/2026
Time Series Causal Discovery via Context-Conditioned and Causality-Augmented Pretraining
PTCD proposes a pretraining framework for causal discovery in time series data with transfer capabilities across diverse domains.
6/15/2026
Traditional machine learning vs. deep learning from dynamic graph representations of proteins' 3D folds in the task of protein structure classification
Compares traditional ML and deep learning approaches for protein structure classification using dynamic graph representations of 3D folds.
6/15/2026
Silent Failures in Federated Personalization of Foundation Models
Study identifies 'silent failures' in federated learning of foundation models, including bias amplification and alignment erosion during personalization.
6/15/2026
OmniOPD: Logit-Free On-Policy Distillation via Speculative Verification
OmniOPD improves on-policy distillation for training student LLMs without requiring teacher logits, addressing distribution shift and credit assignment problems.
6/15/2026
Multi-component Causal Tracing in Large Language Models
Unified framework for multi-component causal tracing in LLMs to identify causal pathways linking inputs to model behavior.
6/15/2026
Adaptive Oscillatory-State Alignment for Time Series Forecasting
Forecasting method for long-term time series using adaptive oscillatory-state alignment for non-rigid periodicity.
6/15/2026
Holding the FP8 Quality Ceiling at 8-Bit Weights and Activations: INT8 and GGUF Post-Training Quantization of Ideogram 4.0 for Consumer GPUs
Post-training quantization of Ideogram 4.0 diffusion transformer to INT8 and GGUF formats for consumer GPU deployment.
6/15/2026
Anatomy of Post-Training: Using Interpretability to Characterize Data and Shape the Learning Signal
Interpretability-based analysis of language model post-training examining data quality and reward signal design impact.
6/15/2026
Quickest Detection of Hallucination Onset: Delay Bounds and Learned CUSUM Statistics
Quickest change detection approach for detecting hallucination onset in LLM token streams using CUSUM statistics.
6/15/2026
Let's Ask Gauss: Improved One-Run Privacy Auditing
Efficient one-run privacy auditing method for differentially private machine learning using Gaussian statistics.
6/15/2026
DeepJEB++: Foundation Model-Driven Large-Scale 3D Engineering Dataset via 2D Latent Space Augmentation
DeepJEB++ foundation model for generating large-scale paired 3D engineering geometry and physics-based performance datasets.
6/15/2026
TWLA: Achieving Ternary Weights and Low-Bit Activations for LLMs via Post-Training Quantization
Post-training quantization method achieving ternary weights and low-bit activations for LLM compression and deployment.
6/15/2026
MP3: Multi-Period Pattern Pre-training for Spatio-Temporal Forecasting
MP3 pre-training method for spatio-temporal forecasting addressing temporal mirage problem in graph neural networks.
6/15/2026
From Uncertain Judgments to Calibrated Rankings: Conformal Elo Estimation for LLM Evaluation
Conformal Elo estimation framework for calibrating LLM-as-a-judge rankings while accounting for systematic biases and uncertainty.
6/15/2026
Quantizing Time-Series Models As Dynamical Systems: Trajectory-Based Quantization Sensitivity Score
Trajectory-based Quantization Sensitivity Score metric for post-training quantization using dynamical systems analysis.
6/15/2026
Dense Supervision, Sparse Updates: On the Sparsity and Geometry of On-Policy Distillation
Analysis of on-policy distillation in language and vision-language models examining sparsity and parameter geometry of updates.
6/15/2026
Application of Artificial Intelligence and Machine Learning in Libraries: A Systematic Review
Systematic review of AI and machine learning applications in library systems and information management.
6/15/2026
On Rate-Optimal Partitioning Classification from Observable and from Privatised Data
Convergence rate analysis for partitioning classification under relaxed conditions with privacy considerations.
6/15/2026
Arbitrary control over multimode wave propagation for machine learning
Photonic processor device with controlled multimode wave propagation for computation through photonic interference patterns.
6/15/2026
Stability of a Generalized Debiased Lasso with Applications to Resampling-Based Variable Selection
Theoretical analysis of generalized debiased Lasso estimator with stability principles and variable selection applications.
6/15/2026
MirrorCheck: Efficient Adversarial Defense for Vision-Language Models
MirrorCheck framework detects adversarial attacks on vision-language models using text-to-image regeneration in multimodal settings.
6/15/2026
Automatic identification of diagnosis from hospital discharge letters via weakly supervised Natural Language Processing
Weakly supervised NLP pipeline for classifying diagnoses in Italian hospital discharge letters without manual document-level annotation.
6/15/2026
ANSR-DT: A Neuro-Symbolic Framework for Adaptive and Explainable Digital Twins
Neuro-symbolic framework combining anomaly detection, symbolic reasoning, and RL for interpretable industrial digital twins.