· ai
[Paper] Procedural Fairness in Multi-Agent Bandits
In the context of multi-agent multi-armed bandits (MA-MAB), fairness is often reduced to outcomes: maximizing welfare, reducing inequality, or balancing utiliti...
-
- · ai
[Paper] Action100M: A Large-scale Video Action Dataset
Inferring physical actions from visual observations is a fundamental capability for advancing machine intelligence in the physical world. Achieving this require...
- · ai
[Paper] Be Your Own Red Teamer: Safety Alignment via Self-Play and Reflective Experience Replay
Large Language Models (LLMs) have achieved remarkable capabilities but remain vulnerable to adversarial ``jailbreak'' attacks designed to bypass safety guardrai...
- · devops
[Paper] Mitigating GIL Bottlenecks in Edge AI Systems
Deploying Python based AI agents on resource-constrained edge devices presents a runtime optimization challenge: high thread counts are needed to mask I/O laten...
- · ai
[Paper] Form and Meaning in Intrinsic Multilingual Evaluations
Intrinsic evaluation metrics for conditional language models, such as perplexity or bits-per-character, are widely used in both mono- and multilingual settings....
- · ai
[Paper] Representation-Aware Unlearning via Activation Signatures: From Suppression to Knowledge-Signature Erasure
Selective knowledge erasure from LLMs is critical for GDPR compliance and model safety, yet current unlearning methods conflate behavioral suppression with true...
- · ai
[Paper] Learning Latency-Aware Orchestration for Parallel Multi-Agent Systems
Multi-agent systems (MAS) enable complex reasoning by coordinating multiple agents, but often incur high inference latency due to multi-step execution and repea...
- · devops
[Paper] Chebyshev Accelerated Subspsace Eigensolver for Pseudo-hermitian Hamiltonians
Studying the optoelectronic structure of materials can require the computation of up to several thousands of the smallest eigenpairs of a pseudo-hermitian Hamil...
- · ai
[Paper] Defending Large Language Models Against Jailbreak Attacks via In-Decoding Safety-Awareness Probing
Large language models (LLMs) have achieved impressive performance across natural language tasks and are increasingly deployed in real-world applications. Despit...
- · ai
[Paper] Model See, Model Do? Exposure-Aware Evaluation of Bug-vs-Fix Preference in Code LLMs
Large language models are increasingly used for code generation and debugging, but their outputs can still contain bugs, that originate from training data. Dist...
- · ai
[Paper] Agent Skills in the Wild: An Empirical Study of Security Vulnerabilities at Scale
The rise of AI agent frameworks has introduced agent skills, modular packages containing instructions and executable code that dynamically extend agent capabili...
- · devops
[Paper] SCRamble: Adaptive Decentralized Overlay Construction for Blockchain Networks
Despite being under development for over 15 years, transaction throughput remains one of the key challenges confronting blockchains, which typically has a cap o...