Category
6,756 papers
Surprisingly, like humans but to a greater extent, LLMs also demonstrated the same hypocrisy in condemning others who, like them, had deployed Bayes'rule.
The results indicate that any immediate effects on opinions of AI are modest, although they leave open the possibility that anthropomorphic language could have an effect in naturalistic settings, or over gradual, continued exposure.
It is found that agents are less effective at resolving tasks when building on agent code compared to human code, with task resolve rate drops of up to 13.1%.
CHERRL is introduced, a Controllable Hacking Environment for Rubric-based RL that enables stable reproduction of reward hacking, explicit observation of reward divergence, and identification of hacking onset and analyzes different judge biases from the perspectives of discoverability and exploitability.
Reach audiences
Advertise in front of researchers, engineers, and readers.
An algorithm is introduced that uses the base model's next-token entropy as a proxy to identify key decision points and resample from those positions and it is proved that the method's mixing time scales with the number of decisions in a trace rather than with the number of tokens, which can be much larger.
This work introduces a temporal-graph-learning (TGL) controller that learns these predictions jointly and supplies both outputs in one forward pass, and shows that a small graph model can handle both decisions and improve the language agents it controls.
The framework and EvalAwareBench provide the tools to measure, attribute, and mitigate evaluation awareness, building the foundation for future solutions to ground evaluation awareness in social psychology.
PRISM, a closed-form upper bound on the cross-entropy risk gap between a target model and a proxy variant, is derived and it is proved that it splits exactly into three measurable axes: scale, shape, and head.
Adaptive optimizers such as Adam are standard for training Transformers, but storing gradient first and second moments incurs substantial memory overhead. We introduce PowerStep, a memory-efficient optimizer that achieves coordinate-wise adaptivity without storing second-moment statistics. Motivated by $\ell_p$-norm st...
GRAVITY (Generation-time Relational Anchoring Via Injected Topological MemorY), a host-independent auxiliary memory layer, is introduced and shows that generation-time structure is a portable complement to existing memory retrieval, while its interaction with host evidence depends on the benchmark and host.
Chain-of-thought (CoT) distillation transfers reasoning behaviors from a strong teacher to a smaller student, but prior work reports a capacity gap: distillation may fail when the teacher-student capability mismatch is large. We revisit the capacity gap from a practical perspective by re-examining commonly used experim...
Screening is introduced, whose explicit threshold transforms bounded query--key similarities into relevance values, enabling exact rejection, empty selection, and direct inspection on a common scale, in a controlled comparison of 12 attention mechanisms on a matched Transformer backbone.
A new language-processing tool could help identify the highest-risk individuals from natural language, enabling swifter interventions.
A Blog post by Liquid AI on Hugging Face
Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking are our most advanced live dialogue models yet, built for natural conversation.
A new method, called CW-Net, translates the reasoning process of an autonomous vehicle’s AI system into understandable concepts that explain its behavior.
With millions of users across the world, Julia has been used to conduct cutting-edge research and to design new drugs, jet engines, heat pumps, and more.
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
