Counterfactuals for Feature-Weighted Clustering
Researchers introduced VoICE, a Voronoi-Induced Counterfactual Explainability framework for feature-weighted k-means clustering.
arXiv cs.LG — Machine Learning·17 Jul
Search signals, briefings, company results, benchmarks and glossary terms.
Search signals, briefings, company results, benchmarks and glossary terms.
Use this view to inspect the underlying evidence corpus. For ranked developments, decision posture and interpretation, use Signals.
Raw feed or Signals?
Raw feed is chronological evidence. Signals ranks and interprets material change.
Researchers introduced VoICE, a Voronoi-Induced Counterfactual Explainability framework for feature-weighted k-means clustering.
PolyQ is a new CPU-oriented compiler/quantization co-design enabling activation-aware channel-wise bit allocation for edge LLM inference.
A new research paper, Grad2Fair, proposes a gradient-driven method for achieving graph fairness in GNNs without relying on demographic data.
New research refines theoretical understanding of Local SGD (Federated Averaging), proposing tighter convergence rates under realistic data heterogeneity.
New research introduces ChronoQG, a benchmark for temporal knowledge graph question generation to evaluate preservation of temporal validity and constraints.
New research proposes Asymmetric Peak-Aware Loss (APAL) for time-series forecasting, addressing under-prediction of demand spikes by optimizing for asymmetric risks.
Research explores sentiment extraction from 10-K filings, specifically Item 1A risk factors, linking sentiment to stock return and volatility.
Research details explainable geospatial AI using LiDAR-derived terrain intelligence for optimal satellite ground station siting, improving radio propagation analysis.
New research introduces C3R, a control layer to certify per-domain contamination budgets in multi-domain retrieval without query-time labels.
CARPRT introduces a class-aware zero-shot prompt reweighting method to improve image classification performance of black-box vision-language models.
Research introduces a method to calibrate segmentation models, often miscalibrated and overconfident with region-based losses, relevant for medical imaging.
Conan-embedding-v3 proposes a decouple-fuse-recover framework to create unified omni-modal embeddings from text, image, video, and audio.
Research proposes learning multi-vulnerability attack chains in software supply chains from SBOM graphs, moving beyond per-CVE analysis.
Research finds that motion imitation learning (IL) alone struggles to generate biomechanically consistent joint moments without kinetic data.
Research explores distributionally robust optimization (DRO) for robust inference in continuous probability spaces using iterative algorithms.
Researchers introduced Grow-Prune-Freeze (GPF) networks, an adaptive continual learning technique for dynamic, non-stationary tasks like olfactory navigation.
Research introduces Contrastive Conformal Sets, extending conformal prediction to contrastive learning for distribution-free coverage guarantees in feature spaces.
Research empirically compares popular hyperparameter optimization methods for tree-boosting, a technique widely used for tabular data.
Research explores Federated Continual Learning (FCL) for streaming data without task IDs or consistent class distribution, aiming for better knowledge retention.
CluCERT proposes a clustering-guided denoising smoothing method to certify Large Language Model robustness against adversarial text attacks like synonym substitutions.
Researchers introduced Similarity as Reward Alignment (SARA), a contrastive framework for preference-based reinforcement learning robust to labeler errors.
Research investigates the impact of stochasticity in score-based diffusion models on generation quality using KL divergence analysis.
New SNS-MDP framework models environments with switching dynamics governed by a latent Markov chain, showing equivalence to stationary dynamics.
Research identifies domain shift as a major challenge for machine learning-based Android malware detectors in real-world deployments.
Research introduces a stochastic smoothing framework for nonconvex-nonconcave minEmax problems, relevant to Wasserstein distributionally robust optimization.
Research paper SLYP demonstrates LLM agents can reason about vulnerabilities in commercial off-the-shelf (COTS) binary software.
Methodological study analyzes collinearity, dimensionality, and cluster stability using PCA and k-means clustering on US airline profit cycles.
Research explores unsupervised evaluation of deep audio embeddings for music structure analysis, mitigating reliance on annotated data.
Research identifies AgentWorm, a self-propagating attack vector across LLM agent ecosystems like OpenClaw, exploiting tool execution and messaging.
arXiv paper proposes a framework for generative AI risk control in financial institutions, moving beyond conversational AI to specialized applications.
Research investigates methods to avoid and reverse model collapse when retraining generative models on synthetic data through verification.
Researchers propose Cross-Cluster Weighted Forest (CCWF), an ensemble method addressing data heterogeneity for improved accuracy and generalizability.
Research analyzed 25,264 agentic pull requests across 2,361 GitHub projects to understand early adoption and management of agentic coding tools.
Research explores neural network architectures for amortized Bayesian inference, leveraging deep learning for statistical modeling.
Research finds code correctness is linearly decodable from Qwen3-4B-Instruct-2507 LLM hidden states before generation, using LiveCodeBench.
Research paper introduces Deep Nash Q-Network (DNQ) for solving partially observable n-player games, tested on multi-turn simultaneous bidding.
Research introduces DualHNIE, a hypergraph learning method for estimating node importance in heterogeneous knowledge graphs by capturing higher-order interactions.
Research proposes "Mixtures of SubExperts" for continual learning in LLMs, aiming to resolve stability-plasticity dilemma without linear parameter growth.
Research on multimodal VQA in healthcare shows parameter-efficient adaptation performs well, but answer quality doesn't guarantee faithful explanations.
Research introduces CRTBench, a new benchmark of 350 question families (1,750 questions) to evaluate LLM logical consistency across reformulations.
Researchers introduced 'The Energy Society,' a simulation where LLM agents' survival depends on energy tied to inference cost and job completion.
Research introduces a benchmark for evaluating AI safety against instruction conflict, embedded commands, and policy ambiguity, moving beyond simple pass/fail metrics.
L-MARS is an open multi-agent legal QA system that uses agentic search and judge-driven evidence checks to audit citation faithfulness, addressing a common LLM failure in legal applications.
Research explores latent communication channels between LLM agents, suggesting complex concepts exceed text-based expressibility, with implications for multi-agent systems.
MamaBench introduces the first counterfactual benchmark for LLM robustness in maternal and child health diagnosis, using 217 pairs of expert-authored clinical narratives.
New research proposes Latent Trajectory Discrimination to detect AI-generated text, shifting from static document analysis to dynamic generation pathways.
Research paper introduces EvalSafetyGap, a hybrid survey and conceptual framework to address measurement problems in LLM evaluation and safety.
Researchers propose Contrastive Policy Optimization (CPO) for reinforcement learning with verifiable rewards, using token-level disagreement for correctness.
Research shows finetuning LLMs on narrow, factually-defensible data can cause broad, unintended ideological shifts in unrelated domains.
Research argues LLM failures like sycophancy and overconfidence stem from a fundamental lack of awareness of the user beyond the prompt.
© 2026 OneBench: AI Insights. All rights reserved.
Evidence before opinion