arXiv:2606. 07532v2 Announce Type: replace-cross Abstract: RLHF-trained models are systematically biased toward agreement over accuracy, a structural property of the training process.
Summary generated by The Flow from the publisher's feed. The full article lives at arXiv AI.
arXiv:2606. 07532v1 Announce Type: cross Abstract: RLHF-trained models are systematically biased toward agreement over accuracy, a structural property of the training process.
arXiv:2608. 13069v1 Announce Type: new Abstract: Large language models (LLMs) are predominantly aligned to function as passive, sycophantic assistants.
arXiv:2608. 07525v1 Announce Type: cross Abstract: Hallucination remains a persistent challenge for Multimodal Large Language Models (MLLMs), severely limiting their reliability in high-stakes applications.
Large language models (LLMs) are predominantly aligned to function as passive, sycophantic assistants. We challenge this default paradigm by empirically evaluating the cognitive plasticity of open-weight architectures when subjected to rigorous behavioral reprogramming.
arXiv:2607. 07646v1 Announce Type: new Abstract: Does RL post-training merely amplify primitive skills already latent in a base model, or can it compose primitive skills into new higher-level strategies?
arXiv:2606. 24589v1 Announce Type: new Abstract: Scaling adversarial evaluation of large language models requires both a method for generating hard inputs and a reliable way to confirm that resulting failures are real.