Search Results for author:
Found 11 papers, 1 papers with code
Negative Preference Optimization: From Catastrophic Collapse to Effective Unlearning
LLM unlearning aims to eliminate the influence of undesirable data from the pre-trained model while preserving the model's utilities on other tasks.
Is Offline Decision Making Possible with Only Few Samples? Reliable Decisions in Data-Starved Bandits via Trust Region Enhancement
Surprisingly, in this work, we demonstrate that even in such a data-starved setting it may still be possible to find a policy competitive with the optimal one.
In-Context Learning of a Linear Transformer Block: Benefits of the MLP Component and One-Step GD Initialization
We study the \emph{in-context learning} (ICL) ability of a \emph{Linear Transformer Block} (LTB) that combines a linear attention component and a linear multi-layer perceptron (MLP) component.
AutoPRM: Automating Procedural Supervision for Multi-Step Reasoning via Controllable Question Decomposition
Recent advancements in large language models (LLMs) have shown promise in multi-step reasoning tasks, yet their reliance on extensive manual labeling to provide procedural feedback remains a significant impediment.
Spreeze: High-Throughput Parallel Reinforcement Learning Framework
While existing parallel RL frameworks encompass a variety of RL algorithms and parallelization techniques, the excessively burdensome communication frameworks hinder the attainment of the hardware's limit for final throughput and training effects on a single desktop.
Explicifying Neural Implicit Fields for Efficient Dynamic Human Avatar Modeling via a Neural Explicit Surface
This paper proposes a technique for efficiently modeling dynamic humans by explicifying the implicit neural fields via a Neural Explicit Surface (NES).
Trained Transformers Learn Linear Models In-Context
We show that although gradient flow succeeds at finding a global minimum in this setting, the trained transformer is still brittle under mild covariate shifts.
NDF: Neural Deformable Fields for Dynamic Human Modelling
However, the learned canonical representation is static and the current design of the deformation fields is not able to represent large movements or detailed geometry changes.
Off-Policy Fitted Q-Evaluation with Differentiable Function Approximators: Z-Estimation and Inference Theory
We approach this problem using the Z-estimation theory and establish the following results: The FQE estimation error is asymptotically normal with explicit variance determined jointly by the tangent space of the function class at the ground truth, the reward structure, and the distribution shift due to off-policy learning; The finite-sample FQE error bound is dominated by the same variance term, and it can also be bounded by function class-dependent divergence, which measures how the off-policy distribution shift intertwines with the function approximator.
Optimal Estimation of Off-Policy Policy Gradient via Double Fitted Iteration
Policy gradient (PG) estimation becomes a challenge when we are not allowed to sample with the target policy but only have access to a dataset generated by some unknown behavior policy.
A paradigm system for strong correlation and charge transfer competition
In chemistry and condensed matter physics the solution of simple paradigm systems, such as the hydrogen atom and the uniform electron gas, plays a critical role in understanding electron behaviors and developing electronic structure methods.
Chemical Physics
