The interpretability of complex neural networks remains a critical challenge, especially for models already deployed in high-stakes domains. To address this, we adapted the Feature Selection Layer (FSL) to optimize feature weights post-training, a variant we term post-hoc FSL. Our approach reframes the FSL as a lightwe...
Jo\~ao Kenji Suwa, Jo\~ao Pedro Silveira e Silva, Bruno Iochins Grisci· 0 citations
When an object category is well represented in one domain but scarcely observed in another, adapting a diffusion model to the target can be unreliable. Yet the same source-target relationship may have been observed for other categories. We ask whether this recurring relation can be learned and reused for a new semantic...
Jing Jiang, Jin-Ping Peng, R. Deen et al.· 0 citations
Automated manufacturing requires data-driven decisions under uncertainty and sparse observational records, and unplanned downtime is a dominant cost of getting them wrong. Prescriptive maintenance recommends a maintenance action against an impending failure rather than merely predicting it. Established reference archit...
Sparse attention reads only a small subset of the key-value cache, and how to treat the unread tail remains a key challenge. Discarding the tail biases the output, and sampling it as previous approaches do fails at reads of 5% and below, where the tail likely holds heavy hitters the selector missed. Our key observation...
Hyunwoo Yu, Jongbeom Lee, Hyunbin Kim et al.· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
Mix-policy optimization incorporates off-policy expert data into the reinforcement learning (RL) loop to transcend the capability boundary of on-policy RL. However, existing methods either provide static full-trace guidance that suppresses self-exploration, or adopt expert prefix guidance without correctness guarantees...
Xuyang Zhong, Jiuding Yang, Yiwen Guo et al.· 0 citations
Trained network weights are increasingly treated as data, but there is rarely ground truth about what a population of models should encode, which hampers systematic validation of weight-space methods. This makes parametric dynamical systems a natural test-bed: the dynamics change qualitatively at submanifolds of parame...
Predicting relative trained performance from architectural specifications can guide design when training costs limit experimentation. Neural Spectral Capacity (NSC) provides such a heuristic, but scores projections independently and assigns identical scores to compatible reorderings of the same layer configurations. We...
Faithfully understanding the internal process that occurs as models form their chain of thought is an important problem with the rise in model capabilities. However, this remains challenging due to the variance in sampled responses, making it difficult to isolate the effects of interventions and compare model reasoning...
S. Im, James Oldfield, A. Bibi et al.· 0 citations
Time series anomaly detection (TSAD) is a foundational problem in data mining, aiming to identify observations that deviate significantly from nominal temporal patterns. Driven by escalating data volumes and application complexity, recent research has yielded diverse architectures, including diffusion models, state spa...
Zhijie Zhong, Zhiwen Yu, Pengyu Li et al.· 0 citations
Sparse autoencoders (SAEs) were originally motivated by discovering monosemantic features, a property that characterizes individual features but does not specify what the learned feature set should look like as a whole. As SAEs have increasingly been adopted for attribution, circuit discovery, and intervention, researc...
This work introduces Short-to-Long (S2L), which supplies a long-horizon behavior learner with features trained by smaller-discount source networks, which yields aggregate performance gains, improving the paired terminal median on 40 of 57 Atari games.
Cheng-Quan Tong, B. Ivanovic, Marco Pavone· 0 citations
aggmetrics is a Python package that extends any binary performance measure expressed in the four confusion-matrix counts to multiclass, multilabel, ordinal, cost-sensitive, soft-label and cross-validation settings, by substituting one of four aggregation operators for what libraries usually expose as an average switch....
