A systems engineer argues that the problem of persisting corrections made by experts to large language model (LLM) errors is an operations issue, not ...
A new paper argues that precision, rather than capability, should be the key metric for evaluating AI systems. The author claims that current benchmar...
Researchers have developed a framework called Verifiable Latent Alignments (VLA) to detect and prevent covert coordination in multi-agent communicatio...
Researchers propose SuTRA (Structurally-Unified Tokenization with Root Awareness), an algorithm that preserves morphological structure in tokenization...
Researchers have developed a method called Latent Space Refusal Anchoring (LSR-Anchoring) to help AI models refuse harmful requests in low-resource Af...
Researchers have developed a method to identify an internal 'valence axis' in language models that tracks the emotional tone of text. This axis can be...
Researchers have found that language models (LLMs) exhibit bias when judging their own outputs versus those of others. In a study, ten LLMs were asked...
Researchers have identified a weakness in large language models called abliteration, where an attacker can bypass safety features by using specific pr...
Researchers have developed a multilingual language model called NE-BERT that can process nine Northeast Indian languages, including Hindi and English ...
Researchers have identified vulnerabilities in language models and vision-language models that can be exploited by 'backdoor' attacks. These attacks a...
Researchers have developed a new caching algorithm called Fractional Decay KV-Cache, designed to improve inference relevance in dialog systems. The al...
Researchers have developed the Middle East Cultural Sensitivity Score (MECSS) to measure structural discourse bias in large language models. The score...