Oceania
Zero-Shot Tokenizer Transfer Benjamin Minixhofer
Our method comes close to the original models' performance in cross-lingual and
NAIS-Net: Stable Deep Networks from Non-Autonomous Differential Equations
Marco Ciccone, Marco Gallieri, Jonathan Masci, Christian Osendorfer, Faustino Gomez
Each block represents atime-invariant iterativeprocess as the first layer in thei-th block,xi(1), is unrolled into a pattern-dependent number,Ki, of processing stages, using weight matricesAi andBi. The skip connections from the input,ui, to all layers in blockimake the process nonautonomous. Blocks can be chained together (each block modeling adifferent latent space) by passing final latentrepresentation,xi(Ki),ofblockiastheinputtoblocki+1.
PIXIU: A Large Language Model, Instruction Data and Evaluation Benchmark for Finance
Despite these successes, the highly technical nature of financial texts requires domain-specific LLMs to understand complex financial language and concepts effectively.
Scalable Deep Generative Relational Model with High-Order Node Dependence
Xuhui Fan, Bin Li, Caoyuan Li, Scott SIsson, Ling Chen
Neural Information Processing Systems http://nips.cc/
On the Global Convergence of (Fast) Incremental Expectation Maximization Methods
Belhal Karimi, Hoi-To Wai, Eric Moulines, Marc Lavielle
Neural Information Processing Systems http://nips.cc/
AccurateLayerwiseInterpretableCompetence Estimation
Our contributions are twofold: First, we establish a statistically rigorous definition of competence that generalizesthecommon notion ofclassifier confidence; second, wepresent theALICE (Accurate Layerwise Interpretable Competence Estimation) Score, a pointwise competence estimator foranyclassifier.
Regret Minimization for Reinforcement Learning with Vectorial Feedback and Complex Objectives
Neural Information Processing Systems http://nips.cc/
Transfer of Deep Reactive Policies for MDP Planning
Aniket (Nick) Bajpai, Sankalp Garg, None
Neural Information Processing Systems http://nips.cc/