Talos is an open-source tool that automates iterative reanalysis of genomic data to identify rare disease diagnoses. It achieved a 90% recovery rate of in-scope diagnoses with only 1.3 candidate variants per patient, and delivered 241 new diagnoses across 5,000 undiagnosed patients, with most new findings emerging within 32 days of evidence publication.
A small-scale experiment shows that native binary embedding models achieve better retrieval than post-hoc binarization of float models. At SciFact Recall@10, native binary models (2048-dim and 4096-dim) outperform post-hoc binary models by 17% and 25% respectively, with significant speed and memory advantages in indexing.
OpenBioRQ introduces a benchmark of 12,553 unsolved biomedical research questions across 12 domains, designed to test agentic models' faithfulness and abstention. It evaluates models in a tool-using setting without answer keys, using real follow-up evidence rather than parametric knowledge, and reveals significant agentic collapse on the hardest questions where tools are no longer used despite being critical.
Mateusz has developed a full pre-trained language model, Project Inkblot's Titan v1, combining Mamba SSM, Multi-Head Attention, and 32-expert MoE in a single decoder-only architecture under 1B parameters. The model, trained on a single NVIDIA L4 GPU for ~$50, achieves 27.5 validation perplexity and demonstrates efficient scaling via a single-line config update, with all components implemented from scratch in PyTorch. Titan v2's first training cycle is now complete, and dataset expansion is underway.
A new dataset, IFLLM, collects mouse trajectories and eye gazing data from users interacting with LLMs. It shows that implicit feedback significantly improves LLM alignment, boosting text-based reward model accuracy from 55% to 64% and nearly tripling response quality improvements after DPO training on eight LLMs.
A new dataset, IFLLM, collects mouse trajectories and eye gazing data from users interacting with LLMs. It shows that implicit feedback significantly improves LLM alignment, boosting text-based reward model accuracy from 55% to 64% and nearly tripling response quality improvements after DPO training on eight LLMs.
ScaffoldAgent introduces a utility-guided framework for dynamic outline optimization in open-ended deep research. It models outline evolution through Expansion, Contraction, and Revision operations, guided by a feedback mechanism that evaluates retrieval gain, structural coherence, and generation quality. Experiments show it improves long-form report generation and factual grounding compared to existing agents.
As enterprise agent tool catalogs scale from 10 to 110 agents, routing accuracy drops 16--23 percentage points on under-specified requests. An oracle analysis identifies retrieval and confusion gaps, with embedding-based shortlisting recovering +10--11pp F1. A human-annotated study of 1,435 utterances confirms real-world recovery of +10--17pp despite lower absolute performance.
Sparse Mixture-of-Experts (MoE) architectures often struggle with low-resource languages due to cross-lingual routing divergence that limits expert sharing. To address this, researchers propose SARA, a framework that transfers specialized capabilities from high-resource anchor languages to low-resource ones. SARA aligns the internal routing distributions of MoE layers using a symmetric Jensen-Shannon divergence constraint rather than operating on output logits. This approach encourages mechanistic consistency in expert selection across different languages. The authors evaluated the method on two large language models across five low-resource languages and three benchmarks. Results show SARA outperforms standard instruction tuning, achieving gains of +0.8% on Qwen3-30B-A3B and +1.2% on Phi-3.5-MoE-instruct for Global-MMLU. These findings demonstrate that SARA effectively addresses performance bottlenecks in low-resource contexts.
NVIDIA's NeMo AutoModel enables faster fine-tuning of transformer models by automating model selection and optimization. It reduces development time and improves efficiency in training large language models on NVIDIA hardware.
Large language models learn causal structure through a difference-making logic, akin to the experimental method. This approach identifies which word sequences influence outcomes and which do not, using vast text data during training. Architectural features like token embeddings and self-attention support this inductive process by detecting patterns of variation and indifference in language.
A fully differentiable emulator of the Atari 2600 VCS is presented, reproducing all 64 ALE games with bit-for-bit accuracy in RAM and screen output. The system enables gradient-based explainable AI by providing a complex, fully known ground truth, with both Julia and JAX implementations validated against a reference emulator and capable of high-throughput differentiable rollouts on GPU.
The paper proposes a self-evolving cognitive framework that uses causal world modeling to enable embodied systems to continuously refine their internal models through interaction. It integrates causal modeling, intervention-driven reasoning, and continual refinement, redefining embodied interaction as an epistemic process for causal discovery and knowledge acquisition. The framework supports a shift from predictive to epistemic intelligence, with a new benchmark for evaluating self-evolving embodied scientific intelligence.
PRIME introduces a framework to analyze how large language models handle conflicting instructions by generating calibrated conflicts in response length, format, and reasoning. The study finds that conflict type has a greater impact on model behavior than model size, revealing diverse failure modes across conflict categories. Results highlight the need for conflict awareness and suggest instruction following cannot be reliably assessed through isolated benchmarks alone.
FACTOR introduces an inference-time model that adapts verification criteria based on claim-level uncertainty. It improves factuality and reduces verification cost by dynamically allocating effort to high-risk claims, demonstrating effective and model-agnostic performance on the FactScore benchmark.
VADAOrchestra introduces a neurosymbolic framework that combines LLM-based workflow orchestration with Datalog+/- symbolic reasoning. It enables adaptive, explainable decision-making by incrementally planning workflows and executing logical inference on demand, offering verifiable traces, auditability, and scalability over large datasets.
SCOPE introduces a framework that refines action plans and evolves symbolic world models in open-ended environments. It combines a Symbolic Execution Simulator and a Self-Adaptive Symbolic Memory to improve plan completeness, perturbation resilience, and cross-task adaptability.
Agentic AI performance degrades exponentially in non-deterministic environments, with k-step success falling as δ^k when per-step determinism δ < 1. The paper introduces a framework linking environment determinism to task success, verifiability, and skill evolution, proposing a Supply Certainty Index and a five-level Determinism Maturity Model. It challenges prevailing views by identifying determinism as a binding constraint across compute, data, embodiment, and alignment.
CCPL introduces a lightweight framework that anchors class prompts to frozen concept prototypes, improving few-shot CLIP adaptation. It achieves better base-to-new performance on DTD and EuroSAT compared to CoOp, with consistent gains from text-space concept regularization, while maintaining neutrality on OxfordPets. The method uses concept dropout and controllable ensemble fusion at inference, with results sensitive to dataset semantics and protocol.
A new Text2DSL system uses context-aware distillation with a structured context of BNF grammar, API specification, and closed identifier vocabulary. Ablation studies show that the vocabulary has the largest impact on semantic quality, while API and BNF significantly improve structural validity, confirming structured context as a critical, load-bearing component.