Real-time AI-curated news from 106,498+ articles across 50+ sources. Sentiment analysis, importance scoring, and key takeaways — updated every 15 minutes.
Researchers propose FSPO (Few-Shot Preference Optimization), a meta-learning algorithm that personalizes large language models using minimal user preference data. The approach uses synthetically generated preferences to train models that can quickly adapt to individual user preferences, achieving 87% performance on synthetic users and 70% on real human users in evaluation tasks.
Researchers propose FedTSP, a federated learning method that uses pre-trained language models to generate semantically-enriched prototypes for improving model performance across heterogeneous data. The approach leverages textual descriptions of classes to preserve semantic relationships while mitigating data heterogeneity challenges in federated settings.
A new study reveals that using large language models to generate synthetic datasets ("silicon samples") produces highly variable results depending on configuration choices, with correlation outcomes ranging from r=.23 to r=.84 on the same task. This demonstrates that analytic flexibility in LLM-based data generation poses a significant threat to research validity and reproducibility in social science applications.
Researchers have developed an intelligent healthcare imaging platform using Vision-Language Models (VLMs), specifically Google Gemini 2.5 Flash, to automate medical image analysis and clinical report generation across CT, MRI, X-ray, and ultrasound modalities. The system achieves 80-pixel average deviation in location measurement and demonstrates zero-shot learning capabilities, though the authors acknowledge clinical validation is necessary before widespread adoption.
Researchers introduced RoleConflictBench, a benchmark dataset containing over 13,000 scenarios across 65 social roles designed to test whether large language models prioritize contextual cues or learned preferences when facing conflicting role expectations. Analysis of 10 leading LLMs revealed that models predominantly rely on ingrained role preferences rather than responding dynamically to situational urgency, indicating a significant gap in contextual sensitivity.
Researchers propose Adaptive Entropy Regularization (AER), a dynamic framework that addresses policy entropy collapse in LLM reinforcement learning by adjusting exploration intensity based on task difficulty. The method improves upon fixed entropy regularization approaches, demonstrating consistent gains in mathematical reasoning benchmarks while maintaining balanced exploration-exploitation tradeoffs.
Researchers have created the first comprehensive Arabic Cultural QA benchmark that translates questions across Modern Standard Arabic and regional dialects, converting multiple-choice questions into open-ended formats. Testing reveals that large language models significantly underperform on dialectal content and struggle with open-ended Arabic questions, highlighting critical gaps in culturally grounded language understanding.
Researchers formalize the one-sided conversation problem (1SC), where only one participant's dialogue can be recorded—common in telemedicine, call centers, and smart glasses. The study evaluates methods to reconstruct missing speaker turns and generate summaries from incomplete transcripts, finding that smaller models require finetuning while larger models show promise with prompting techniques.
Researchers introduce MTR-DuplexBench, a new evaluation framework for Full-Duplex Speech Language Models that enables real-time overlapping conversations. The benchmark addresses critical gaps by assessing multi-round interactions across conversational quality, instruction-following, and safety dimensions, revealing that current FD-SLMs struggle with consistency across multiple communication rounds.
Researchers demonstrate that LLMs can be used as lossless encoders and decoders for invertible problems in hardware design, significantly reducing hallucinations and omissions. By generating HDL code from Logic Condition Tables and reconstructing the original tables to verify accuracy, the approach improves developer productivity and catches both AI-generated errors and design specification flaws.
Researchers propose TPA (Token Probability Attribution), a new method for detecting hallucinations in Retrieval-Augmented Generation systems by attributing token generation to seven distinct sources rather than the traditional binary approach. The technique uses Part-of-Speech tagging to identify anomalies in how different linguistic categories are generated, achieving state-of-the-art detection performance.
Researchers have introduced VLegal-Bench, the first comprehensive benchmark for evaluating large language models on Vietnamese legal tasks, comprising 10,450 expert-annotated samples grounded in real legal documents. The benchmark uses Bloom's cognitive taxonomy to assess LLM performance across practical legal scenarios, establishing a standardized framework for developing more reliable AI-assisted legal systems in Vietnam.
Researchers identify specific attention heads in vision-language models that cause prompt-induced hallucinations, where models favor textual instructions over visual evidence. By ablating these identified heads, they reduce hallucinations by 40% without retraining, revealing model-specific mechanisms underlying this failure mode.
Researchers propose VIB-Probe, a novel framework using Variational Information Bottleneck theory to detect and mitigate hallucinations in Vision-Language Models by analyzing internal attention mechanisms. The method identifies specific attention heads responsible for truthful generation and introduces an inference-time intervention strategy that outperforms existing detection baselines.
EnvScaler is an automated framework that generates synthetic tool-interaction environments for training LLM agents through programmatic synthesis, creating 191 diverse environments and 7,000 scenarios. The approach addresses scalability challenges in LLM agent training by combining topic mining and logic modeling to overcome hallucinations and manual bottlenecks, demonstrating improved performance on multi-turn, multi-tool interaction tasks.
Researchers introduce CLewR, a curriculum learning strategy that improves machine translation performance in large language models by reordering training data from easy to hard examples with periodic restarts. The approach demonstrates consistent improvements across multiple model families and preference optimization techniques, addressing a previously underexplored aspect of LLM training methodology.
Researchers introduce Evolve-CTF, a tool that generates families of semantically-equivalent cybersecurity challenges to evaluate the robustness of agentic LLMs. Testing 13 LLM configurations reveals models are resilient to basic code transformations but struggle with obfuscation and composed modifications, providing new benchmarking methodology for AI safety evaluation.
Barclays has pushed back its forecast for ECB rate hikes to June and September 2026, signaling expectations of persistent inflation pressures in the Eurozone. This delay could introduce market volatility and affect cryptocurrency valuations tied to macroeconomic conditions.
China's diplomatic support for Iran is complicating US-Iran relations and potentially delaying a planned Trump visit by April 30. The geopolitical tension between the US, China, and Iran creates uncertainty in international relations that could have downstream effects on global markets and cryptocurrency trading patterns.
Trump's decision to avoid military deployment on Kharg Island signals a pivot toward economic pressure rather than military intervention in regional conflicts. This geopolitical restraint reduces immediate military escalation risks, potentially stabilizing markets that had priced in conflict premiums, particularly affecting oil prices and risk assets.
Ethereum has declined below the $2,350 level and is consolidating near $2,250, with technical indicators showing bearish momentum. The price faces key resistance at $2,360 and $2,415, while major support sits at $2,250 and $2,220, creating a critical decision point for the near-term trend direction.
Venture capital funding for crypto startups has become significantly more difficult to secure, with higher scrutiny and stricter criteria now applied by top-tier VCs. This represents a marked shift from the easier funding environment of previous cycles, reflecting broader caution in the crypto investment landscape.
Bitcoin retreated from $78,400 to trade below $75,500, testing key support levels as technical indicators turn bearish. While the overall structure remains potentially bullish, BTC faces critical resistance at $75,500 and risks further decline if support at $73,500 fails to hold.
Improving US-Iran diplomatic relations are creating conditions where Bitcoin and alternative assets may gain adoption as investors hedge against dollar volatility. The narrowing geopolitical tensions, combined with broader currency concerns, position cryptocurrency as a potential safe haven for traders seeking diversification away from traditional fiat exposure.
Iran has accused the United States of breaching diplomatic commitments, escalating tensions between the two nations and undermining ongoing ceasefire negotiations. The impasse threatens to destabilize the region and dim prospects for meaningful US-Iran dialogue.
