Safety & Evaluation
A
Showing 31–60 of 70
-
HEARSAYBENCH: Can AI Navigate from Abstract Human Rights to Lived Lives?
-
BASILISC: Bayesian Alignment of Siamese-Integrated Latents for Inferring Single-cell Correspondence
-
How Architecture, Typology, and Adaptation Strategies Shape ASR for Low-Resource African Languages: A Controlled Empirical Study
-
Google、「Gemini 4 Carbon」発表 「Argon」公開から2週間、サイバー防御能力などを強化Google unveils Gemini 4 Carbon with stronger cyber defense and autonomyGoogle announced Gemini 4 Carbon, a frontier AI model with stronger cybersecurity and long-running autonomous work, topping benchmarks. It also resists hallucinations and manipulation better. Access starts with select cyber defense organizations, widening gradually.
-
Discovering Jailbreaks via Curriculum of Models
-
Beyond fabrication: Misattributed numbers in papers written by AI research agents
-
Information-Theoretic Reliability--Safety Bounds and Finite-Alignment Dynamics in AI Knowledge Distillation
-
Curriculum-Based Deep Reinforcement Learning for Robust Autonomous Rocket Landing under Atmospheric Uncertainty
-
Rethinking LLM Jailbreak Evaluation: Continuous Risk Landscapes Beyond Binary Outcomes
-
An Alignment Benchmark for Enterprise Agents on Era by Eon
-
Value Probes: Learned Value Functions for Beam-Search Red Teaming
-
What Hallucinations Have We Measured? Semantic Illusion, Confabulation, and Expressed Uncertainty
-
Accelerating Nonlinear MPC by Behavior Cloning with Separated Timescales: Application to Nuclear Reactor Load-Following
-
PFNの国産フラッグシップモデル「PLaMo 3.1 Prime」発表 オンプレでDeep Research対応 API無料プランも開始PFN launches PLaMo 3.1 Prime, a domestic LLM with on-prem Deep ResearchPreferred Networks has released PLaMo 3.1 Prime, its domestically developed flagship generative AI foundation model. It adds Deep Research that can run on-premises and is said to reduce hallucinations. A free API plan is also available for trying it out.
-
日本取引所、金融市場の取引システム運用監視にAIを導入 精度97％を実現した方法JPX and Hitachi deploy AI for trading system monitoring, 97% accuracyJapan Exchange Group and Hitachi have begun applying AI to operational monitoring of financial market trading systems, aiming to ease reliance on individual experts. The article explains how they handled hallucination risk to reach 97% accuracy.
-
Pushing Dexterity and Safety for In-Hand Object Pose Reaching That Works Remarkably Well
-
Multi-State Linear Attention with Generalized Second-Order Routing
-
When uncertainty quantification fails: structural vulnerabilities and corrections for selective prediction under clinical class imbalance
-
Your Robot’s Safety Functions Already Work. What If the Input Lies?IEEE Spectrum asks what happens if a robot's safety inputs are falsifiedRobot safety functions such as slowing near people depend on inputs like distance readings, position estimates or stop signals. This IEEE Spectrum piece, sponsored by VicOne, asks what happens when those inputs are false.
-
Follow the Gradient: Tracing Safety Degradation Back to Training Data
-
When Visual Evidence Is Insufficient: How Instructions, Post-Training, and Internal Activations Shape Refusal in Multimodal Language Models
-
Constraining the Commit Boundary: A DIO/ODI-Informed Authorization Guardrail for Agentic Entity Resolution
-
Game-Theoretic Solvers with Bayesian Opponent Modeling and Moral Rhetoric Framing for GLEE
-
Same Verdict, Different Readers: Personalizing Fact-Checking Explanations from Relative Preferences
-
Uncovering Why LLMs Fail: From Internal Mechanisms to Adversarial Reasoning Trajectories
-
Aligning Institutional Power and Operational Ability
-
Aligning Large Language Models for Safety in African Languages: A Case Study of Hausa
-
Decision Models as Judges for Hallucination Detection
-
Mining the Nineteenth Century for the Thirteenth: Register-Scale Forced Alignment and HTR Bootstrapping via Legacy Regesta and Antiquarian Editions
-
I-HITL: Inalienable Human-in-the-Loop