Researchers introduced SPEAR (Sandboxed Prompt Engineer with Active Roll-back), a code-augmented agentic optimizer that autonomously rewrites prompts using a Python sandbox for structural error analysis, outperforming ex…
A new study published on arXiv describes "The Daily Dose," an LLM-driven system that automates clinical summarization and clinical-trial identification for radiation oncology. After one month of deployment, a survey of 5…
Researchers have developed RICE-PO, a policy optimization framework that converts retrieval interactions into localized learning signals for training reasoning-based retrieval agents. The framework addresses the credit-a…
Researchers have demonstrated that retrieval-augmented generation (RAG) can be understood as an in-context optimization process, where a single linear self-attention layer performs a gradient-descent step on a unified RA…
A new study reveals that large language models hallucinate on structured knowledge tasks due to systematic internal failures, not random errors. Researchers found that attention mechanisms disproportionately focus on sho…
Researchers have developed a framework for evaluating and adjusting the cultural values of large language models by analyzing their responses to scenario-based dilemmas rather than direct questioning. The method, which u…
Researchers introduced EnterpriseMem-Bench, a multi-turn Text-to-SQL benchmark of 300 sessions and 1,400 queries built from three enterprise domains, to evaluate how five frontier AI models handle memory across conversat…
Researchers have developed a framework that uses large language models to deliver just-in-time adaptive feedback to students by grounding the AI with domain-specific expert knowledge. Deployed in a university course with…
Researchers have developed Token-to-Mask (T2M) remasking, a training-free replacement for Token-to-Token (T2T) editing in discrete masked diffusion language models like LLaDA. The method resets suspected erroneous tokens…
Researchers at an undisclosed institution have developed LURE (Live-Usage Replay Evaluations), a method that constructs more realistic AI safety evaluations by replaying real-world agentic interactions and appending eval…
A new survey reframes alignment tuning for large language models as a data pipeline design problem, decomposing data construction into three stages: response synthesis, preference evaluation, and preference instantiation…
Most SaaS AI agents do not require a vector database for retrieval, as file-based memory combined with 1M-token context windows and tool calls handles the typical use case more efficiently. Anthropic's official "key prim…
Researchers have developed UniMVU, a unified multimodal video understanding framework that uses instruction-aware gating to dynamically balance the importance of different input modalities like video, audio, and depth ma…
Researchers introduced LongAV-Compass, a benchmark for evaluating minute-long audio-visual generation across text, image, and video conditioning modalities. The benchmark contains 284 test cases and assesses over 20 fine…
A new study benchmarked twelve deep learning architectures across four model families—convolutional neural networks, vision transformers, hybrid CNN-transformer backbones, and vision-language models—for multi-disease ret…
Researchers have developed Evi-Steer, a new framework for BiomedCLIP that enables uncertainty-aware fine-tuning of biomedical vision-language models by updating only 0.11% of total parameters. The method uses evidential …
Researchers have developed E³C, a video generation framework that creates egocentric footage by separating persistent 3D scene structure from human-driven dynamics using point cloud-based memory and appearance descriptor…
Researchers have developed RadarSim, a unified differentiable renderer that uses multimodal neural fields to generate high-resolution Doppler radar range images by leveraging camera data. The system addresses the low spa…
Researchers have developed BEAP, a black-box adversarial prompting attack that exploits vulnerabilities in text-to-image diffusion models that have undergone machine unlearning. The attack uses a large language model to …
Researchers have developed DecoupleGen, a method that uses personalized text-to-image diffusion models to generate training images of objects in uncommon contexts, such as a beach ball on a road. The approach aims to cor…