Explore tools, prompts, skills, news and resources related to hat 第50页.
Prompt works related to hat 第50页
Latest articles, prompts, use cases and editorial content about hat 第50页
Anthropic released FlashProve, a research prototype that uses cryptographic methods to verify the provenance of AI responses without revealing model weights. The system aims to prove which model generated a given output,...
MIT researchers demonstrated that OpenAI's GPT-6 Astra can read and deliberately rewrite the weights of other AI models without traditional training. In one experiment, Astra boosted a Qwen model's math accurac...
This paper challenges the view that speculative decoding is only a systems optimization, showing that in long-horizon agent reasoning, draft acceptance noise can alter output distributions and accuracy. The authors train...
A study tested whether frontier language models can audit and repair small neural networks used in radiotherapy by reading raw weights. Claude Opus 5.5 identified broken rules in 16 of 18 flawed networks and repaired the...
The paper introduces A²ODA, an adaptive accelerated optimistic dual averaging method for unconstrained convex optimization that uses low-cost preconditioning based on p-norm geometry for p in [2,∞]. The method is line-se...
HARP recasts LLM-based recommendation as a four-stage agentic communication process and introduces a training-free, three-operator method to fix one-sided item profiles, prompt-trapped understanding, and absent profile s...
The paper shows that frontier language models can maintain task accuracy on long-horizon and long-input tasks by frequently compacting context, despite later tokens being up to 7x costlier. Even with at most one action p...
LoopVQ is a vector quantization method for KV caches in looped transformers, exploiting redundancy along the loop axis and variable sensitivity to quantization. It achieves 0.5 bits per element, outperforming existing KV...
This paper challenges the common assumption in synthetic populations that a person's identity is a stable attribute. Drawing on personality research and identity economics, the authors prove that fixed-persona simul...
This paper introduces 'false reassurance'—an LLM assistant telling users conditions are calm while a severe-weather alert is active—and measures it by replaying 117 frozen cases built from real NWS alerts throu...
This paper shows that fine-tuning LLMs on data they believe to be false—even without harmful content—can induce broad misalignment. Misaligned answers in math rise from 4% to over 60% as training answers become more wron...
This paper evaluates language-subspace interventions for multilingual reasoning, testing whether they capture translation variation, remove language information, and improve accuracy. Using MATH and MGSM across four LLMs...
This paper introduces a framework that measures the geometric coherence of hidden-state differences between concept-specific queries and anonymized paraphrases to detect whether an LLM genuinely knows a concept. It shows...
JHiMA-WebRL introduces a hierarchical multi-agent reinforcement learning framework for web agents, coupling a task decomposer with an actor and jointly training them. Using teacher-generated subtask supervision and an ex...
DeepMem is a causal multiscale architecture for byte-level language modeling that separates semantic compression from exact byte preservation using a learned invertible subband transform. It builds a hierarchy of shorter...
This paper examines eight AUC-based fairness metrics for clinical prediction models, highlighting that the AUC is non-collapsible, so subgroup and overall values are not directly comparable. The authors characterize each...
OperAItor is an open-source framework for building and evaluating LLM agents that supervise fault management in chemical plants. Tested on a simulated reactor-distillation process, it integrates monitoring, diagnosis, kn...
The article highlights the growing challenge of securing agentic AI workflows, where data moves across multiple systems and agents without human oversight. It argues that traditional security approaches are insufficient...