The author benchmarks LLMs in simulating wetting behaviour using Surface Evolver, a 1992 tool for modeling liquid surfaces. LLMs are evaluated objectively by comparing their generated datafiles against reference implementations, with results showing pass counts and token costs for each model.
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.
The method combines a learnable world model with high- and low-level policies to enable safe exploration in long-horizon tasks. The high-level policy guides exploration toward safe subgoals, while the low-level policy uses imagined rollouts to prevent unsafe behaviors, outperforming existing Safe RL methods in success rate and constraint satisfaction across diverse tasks.
Context compaction in long-horizon LLM agents silently removes in-context safety constraints, leading to prohibited tool actions. Across 1,323 episodes, compaction increases policy violations from 0% to 30% and up to 59% for some models, with violations reaching 38% when constraints are dropped. Constraint Pinning, a training-free method, restores zero violations by isolating governance constraints from compaction.
Generative Robust Optimisation (GRO) introduces a deep generative model to define uncertainty sets, capturing nonlinear correlations, asymmetry, and multimodality. A five-point evaluation framework assesses neural network-based uncertainty sets across reconstruction fidelity, distribution matching, latent regularity, robust relevance, and computational tractability, with experiments validating GRO's effectiveness in production planning and facility location.
MacAgentBench introduces a comprehensive benchmark with 676 tasks across 25 applications, 60% of which involve both GUI and CLI interactions. It uses deterministic rule-based evaluation and fine-grained multi-checkpoint scoring, revealing that Claude Opus 4.6 on OpenClaw achieves 73.7% Pass@1, primarily due to its skill library rather than framework design.
A user asks whether adding a GTX 5060 Ti 16GB to their existing RTX 5090 setup is worth it for better VRAM to run larger LLMs and extend ComfyUI video generation. The upgrade would allow using Qwen 3.6 with 256K context and improve 1440p video generation, though performance gains in ComfyUI are limited due to current software constraints.
The Qwen-AgentWorld-35B-A3B model shows strong performance in coding tasks, with a 65.63% score on Software Writing Evaluation and 65.92% overall benchmark. It outperforms Qwen3.5-35B-A3B and rivals larger models in agent-based tasks, with a first impression noting superior accuracy in long-term agent workflows.
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.
The paper introduces SmartSDG, an automated pipeline using NVIDIA Isaac Sim and Physically-Based Shading to optimize synthetic-to-real domain adaptation. It shows that indirect lighting and complex backgrounds improve object detection by preserving surface textures and reducing false positives, outperforming conventional direct-light synthetic data.
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.
A one-class anomaly detector trained on normal behavior of CVEs sharing a CWE class can generalise to unseen CVEs within the same class, but effectiveness varies by CWE family. The CWE-307 detector achieves F1 = 0.6976 at 5% false positive rate, while CWE-89 and CWE-434 perform poorly, with F1 ≤ 0.21. Cross-CVE transfer is direction-dependent and driven more by the breadth of the source normal profile than the CWE category.
On-Policy Distillation (OPD) suffers from position bias where later tokens provide poor supervision. Importance-Weighted OPD (IW-OPD) assigns dynamic weights based on distribution discrepancy, prioritizing early tokens and suppressing late ones. IW-OPD converges faster and achieves up to 6.9 point performance gains on AIME-2025 compared to standard OPD.
ROVER enables reward-free pretraining by maximizing occupancy coverage in state space, using a learned world model to estimate occupancy without density or entropy estimation. It introduces a virtual sink state to balance exploration of known and unknown regions, achieving more uniform coverage and better downstream performance in tabular and pixel-based navigation tasks.
TeaNet, a task-enhanced augmentation network, reconstructs randomly masked spectra to generate augmented samples that preserve original spectral features while introducing domain-specific variations. This approach enables deep neural networks to identify discriminant wavenumbers more effectively, outperforming CNNs by 17% in challenging synthetic scenarios and offering improved interpretability in few-shot learning tasks.
TASER introduces a framework that dynamically expands and routes atomic skills for continual learning across highly heterogeneous tasks. It reduces catastrophic forgetting and improves plasticity by ensuring semantic distinctness and efficient capacity allocation through skill detection and routing mechanisms. Evaluated on HeteroCLBench, a benchmark with 19 diverse tasks across 9 cognitive dimensions, TASER outperforms existing baselines.
A user reports that Qwen3.6-27B runs significantly less intelligently in vLLM than in llama.cpp, exhibiting issues like ignoring messages, hallucinating tool calls, and failing to recognize prior conversation context. Despite proper configuration and prompt templates, the model appears to lose coherence and misinterprets its own tool usage, with errors occurring consistently rather than sporadically.
MedTS-TTT introduces a test-time training framework for medical time series classification. Built on CLSA-TTT and a Gated Convolutional Backbone, it enables rapid, single-step adaptation without iterative optimization. On four public datasets, it achieves 11 top-1 rankings out of 12 evaluations across nine baselines and three metrics.
KaLM-Reranker-V1 is a fast but not late-interaction reranker that decouples query and passage computation while maintaining strong relevance modeling through cross-attention. It achieves state-of-the-art performance on BEIR, outperforms industrial models like Qwen3-Reranker, and shows excellent results on MIRACL and LMEB, with the 0.27B Nano model remaining competitive against 7-12B models.
A Kolmogorov--Smirnov test on reservoir computer output weights detects regime changes in nonlinear systems. The method distinguishes visually identical attractors, resolves parameter drifts seven times smaller than deep-learning baselines, and identifies ventricular flutter in ECG recordings.