A novel unified operator is introduced that combines several regularized RL operators into a general framework that better targets peakier sampling distributions and is named trajectory general mellowmax (TGM), which is shown to identify higher quality, diverse candidates than baselines in both synthetic and real-world tasks.
Marco Jiralerspong, Esther Derman, Danilo Vucetic et al.· 2 citations
The strong association between BrainAGE, vascular risk factors, and post-stroke recovery highlights its potential for personalized prognostic modeling in stroke care.
Vincent Roca, Marc Tommasi, Paul Andrey et al.· NeuroImage· 3 citations
EquiReg formalizes manifold-preferential equivariant functions that exhibit low equivariance error for on-manifold samples and high error for off-manifold ones, thereby guiding sampling toward symmetry-preserving regions of the solution space.
KronSAE is proposed, a design that factorizes the latent space into heads and forms post-latent features as pairwise compositions of lower-dimensional pre-latents using mAND, a differentiable AND-like interaction that imposes a compositional co-activation prior while remaining compatible with standard SAE objectives and variants.
Vadim Kurochkin, Yaroslav Aksenov, Daniil Laptev et al.· 1 citation
Reach audiences
Advertise in front of researchers, engineers, and readers.
An attack is proposed, FAB (Finetuning-activated Adversarial Behaviors), which compromises an LLM via meta-learning techniques that simulate downstream finetuning, explicitly optimizing for the emergence of adversarial behaviors in the finetuned models.
Thibaud Gloaguen, Mark Vero, Robin Staab et al.· 4 citations
CP generalized approximate message passing (CP-GAMP) is developed for incomplete noisy Bayesian CPD and synthetic and image-inpainting experiments show that CP-GAMP substantially reduces runtime relative to variational Bayesian CPD while maintaining competitive reconstruction accuracy.
Bingyang Cheng, Zhongtao Chen, Yichen Jin et al.· 0 citations
A family of premetrics that capture different degrees of structural similarity between graphs are introduced and relate these similarities to generalization, and consequently, the performance of expressive GNNs are related.
The central conclusion is conditional rather than adversarial: cosine's positive-scale invariance is justified when radial variation is nuisance or fixed by the representation contract, but its angular geometry and downstream decision must still be validated.
The results demonstrate that augmenting conventional flow features with temporal information yields consistent gains; binary detection improves by up to 3% in F1 score, while macro-averaged multi-class F1 increases by approximately 27%, with the most significant improvements occurring in attack classes with pronounced temporal signatures.
Majed Luay, S. Layeghy, Niloufar Noorbin et al.· IEEE Access· 21 citations
It is shown that RSPO with appropriate regularizers can substantially improve the length-controlled win rate on AlpacaEval-2 across a range of base models, while also achieving consistently superior performance on Arena-Hard, MT-Bench, ArmoRM, and response diversity.
Xiaohang Tang, Sangwoong Yoon, Seongho Son et al.· 6 citations· ⚡1
Layer-Integrated Memory (LIMe) is introduced, a lightweight extension that leverages existing key-value buffers and learns per-head, per-layer routing weights to integrate representations from previous layers to improve perplexity per FLOP and yield strong gains on synthetic tasks while preserving higher value-vector entropy and token separability.
This paper introduces Alert, a Deep-RL framework that learns trigger functions from any state representation, and proposes Alert+, a simple yet effective variant that consistently outperforms traditional methods in balancing accuracy and delay within an imbalanced misclassification and exponential delay cost setting.
Aurélien Renault, A. Bondu, Antoine Cornuéjols et al.· 2 citations
A weeklong summer workshop brought higher education faculty to campus to explore how AI and machine learning materials can be adapted for their classrooms.
Adaptive AI agents can help make BIM data more machine-readable by navigating IFC models, interpreting inconsistent information, and mapping it to defined standards. In this blog, Alok Rawat shares findings from a real-world pilot in construction workflows. The post Adaptive AI Agents in Construction Workflows appeared first on GPT-Lab.
