Reflective text optimization makes agentic systems better, but the optimization loop is itself a resource-hungry workload whose sequential structure leaves parallel evaluation capacity idle. We extend the open-source GEPA optimizer to propose and evaluate a batch of candidates on each optimization step instead of one c...
Jialin Zhang, Lakshya A Agrawal, Shangyin Tan et al.· 0 citations
As large language models advance rapidly, they are increasingly used to answer medical questions, and many such models are now deployed locally. Once a model is deployed, its behavior is governed mainly by the ethical policy written into its constitution. However, most existing work focuses on factual correctness rathe...
Yibo Yang, Zerui Huang, Haomiao Yang et al.· 0 citations
Cross-modal translation requires models to capture shared semantics across modalities while preserving the fine-grained structure required by the target modality. Yet in real-world scientific domains, paired data are often imperfect, while obtaining clean pairs at scale is costly. Supervised fine-tuning (SFT) directly...
Chaofei Hong, Ziyang Chen, Baochen Wang et al.· 0 citations
The findings reveal that VLMs might be capable of rich, creative, and visually-grounded communication, but their interpretation and production of visual meaning diverges from human conventions in surprising ways, and highlights both the promise and challenges of visual communication between humans and VLMs.
Alexa R. Tartaglini, Zhengxuan Wu, Thaddäus Wiedemer et al.· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
This work builds the two inputs: the model states its knowledge as notes, a small policy voices each note with a calibrated confidence, a reader answers in two voices, and a selector commits or abstains under a prior applied as an exact prior shift so that what the model knows, what the context says and what is true ar...
Perimeter triage faces a latency/semantics trade-off: signature engines are fast and format-safe but blind to unseen attacks, while generative models are semantic but slow and unparseable. We present VX-One, a small multi-head network (100K--10M parameters) with calibrated typed outputs: a binary gate, a closed-set cho...
Agent harnesses have become a central component of modern agentic systems, shaping how models use tools, manage context, and coordinate long-horizon interaction. Realizing the benefits of these harnesses depends on how effectively the underlying model can make use of them, motivating reinforcement learning directly thr...
Fei Bai, Huatong Song, Chuan Hao et al.· 0 citations
Gold's theorem shows that no learner can identify an arbitrary language in a superfinite class from positive presentations alone. We prove a positive result of a different shape. Fix an innate \emph{recursive contrast operator} $R$ mapping any string to a finite set of structurally bounded perturbations of it, and a di...
In 2023, a New York judge sanctioned two attorneys in \emph{Mata v. Avianca} for filing a brief containing hallucinated citations generated by ChatGPT. These failures can be largely caught by database lookups. The harder problem is detecting citations that point to real cases but do not support the propositions for whi...
Low-rank adaptation (LoRA) enables efficient specialization and distribution of large language models through compact adapters. However, untrusted adapters introduce a supply-chain threat: a backdoored adapter can cause a model to generate harmful content, malicious code, political propaganda, or covert advertisements...
Doniyorkhon Obidov, Honggang Yu, Xiaolong Guo et al.· 0 citations
Large Language Models (LLMs) have accelerated drug discovery, particularly in the automated design of antimicrobial peptides (AMPs). However, current validation pipelines for peptide generation models overlook historical precedents showing that certain drugs carry health risks predominantly for individuals with specifi...
Doniyorkhon Obidov, Xiaolong Guo, Yonghui Li et al.· 0 citations
In a dual-use field, hazardous knowledge and the benign expertise around it share vocabulary, concepts, and causal structure. That overlap makes selective refusal hard, and existing methods such as unlearning and refusal training overgeneralize their suppression into the benign side. We argue for conditioning, not supp...
Flint-interactive gives AI agents a language for creating interactive visualizations. Click a chart, filter data, highlight related content, and explore beyond the view the author originally chose. The post Flint-interactive: A language for creating interactive data visualizations appeared first on Microsoft Research.
