Search Results for author:
Found 29 papers, 16 papers with code
Ineq-Comp: Benchmarking Human-Intuitive Compositional Reasoning in Automated Theorem Proving on Inequalities
LLM-based formal proof assistants (e. g., in Lean) hold great promise for automating mathematical discovery.
AdaptMI: Adaptive Skill-based In-context Math Instruction for Small Language Models
In-context learning (ICL) allows a language model to improve its problem-solving capability when provided with suitable information in context.
Goedel-Prover: A Frontier Model for Open-Source Automated Theorem Proving
On the miniF2F benchmark, it achieves a 57. 6% success rate (Pass@32), exceeding the previous best open-source model by 7. 6%.
Rethinking Mixture-of-Agents: Is Mixing Different Large Language Models Beneficial?
We confirm that the MoA performance is rather sensitive to the quality, and mixing different LLMs often lowers the average quality of the models.
On the Limited Generalization Capability of the Implicit Reward Model Induced by Direct Preference Optimization
These findings highlight that DPORM has limited generalization ability and substantiates the integration of an explicit reward model in iterative DPO approaches.
Localize-and-Stitch: Efficient Model Merging via Sparse Task Arithmetic
Our algorithm works in two steps: i) Localization: identify tiny ($1\%$ of the total parameters) localized regions in the finetuned models containing essential skills for the downstream tasks, and ii) Stitching: reintegrate only these essential regions back into the pretrained model for task synergy.
Leveraging Invariant Principle for Heterophilic Graph Structure Distribution Shifts
However, existing works are only devoted to designing better HGNN backbones or architectures for node classification tasks on heterophilic and homophilic graph benchmarks simultaneously, and their analyses of HGNN performance with respect to nodes are only based on the determined data distribution without exploring the effect caused by this structural difference between training and testing nodes.
Regularizing Hidden States Enables Learning Generalizable Reward Model for LLMs
Reward models trained on human preference data have been proven to effectively align Large Language Models (LLMs) with human intent within the framework of reinforcement learning from human feedback (RLHF).
On the Benefits of Over-parameterization for Out-of-Distribution Generalization
We demonstrate that in this scenario, further increasing the model's parameterization can significantly reduce the OOD loss.
A Sober Look at the Robustness of CLIPs to Spurious Features
Large vision language models, such as CLIP, demonstrate impressive robustness to spurious features than single-modal models trained on ImageNet.
Arithmetic Control of LLMs for Diverse User Preferences: Directional Preference Alignment with Multi-Objective Rewards
Additionally, DPA models user preferences as directions (i. e., unit vectors) in the reward space to achieve user-dependent preference control.
The Instinctive Bias: Spurious Images lead to Illusion in MLLMs
In this paper, we identify a typical class of inputs that baffles MLLMs, which consist of images that are highly relevant but inconsistent with answers, causing MLLMs to suffer from visual illusion.
R-Tuning: Instructing Large Language Models to Say `I Don't Know'
This approach is formalized by first identifying the disparity in knowledge encompassed by pre-trained parameters compared to that of instruction tuning data.
Continuous Invariance Learning
To address this challenge, we then propose Continuous Invariance Learning (CIL), which extracts invariant features across continuously indexed domains.
Spurious Feature Diversification Improves Out-of-distribution Generalization
Contrary to the conventional wisdom that focuses on learning invariant features for better OOD performance, our findings suggest that incorporating a large number of diverse spurious features weakens their individual contributions, leading to improved overall OOD generalization performance.
Mitigating the Alignment Tax of RLHF
Building on the analysis and the observation that averaging different layers of the transformer leads to significantly different alignment-forgetting trade-offs, we propose Heterogeneous Model Averaging (HMA) to Heterogeneously find various combination ratios of model layers.
Optimal Sample Selection Through Uncertainty Estimation and Its Application in Deep Learning
Our proposed method, COPS (unCertainty based OPtimal Sub-sampling), is designed to minimize the expected loss of a model trained on subsampled data.
What is Essential for Unseen Goal Generalization of Offline Goal-conditioned RL?
In this paper, we study out-of-distribution (OOD) generalization of offline GCRL both theoretically and empirically to identify factors that are important.
Active Prompting with Chain-of-Thought for Large Language Models
For this purpose, we propose a solution to the key problem of determining which questions are the most important and helpful ones to annotate from a pool of task-specific queries.
Model Agnostic Sample Reweighting for Out-of-Distribution Learning
The overfitting issue is addressed by considering a bilevel formulation to search for the sample reweighting, in which the generalization complexity depends on the search space of sample weights instead of the model size.
Probabilistic Bilevel Coreset Selection
The goal of coreset selection in supervised learning is to produce a weighted subset of data, so that training only on the subset achieves similar performance as training on the entire dataset.
Stable Learning via Sparse Variable Independence
The problem of covariate-shift generalization has attracted intensive research attention.
Particle-based Variational Inference with Preconditioned Functional Gradient Flow
With the popularity of Stein variational gradient descent (SVGD), the focus of particle-based VI algorithms has been on the properties of functions in Reproducing Kernel Hilbert Space (RKHS) to approximate the gradient flow.
Self-Guided Noise-Free Data Generation for Efficient Zero-Shot Learning
In this paradigm, the synthesized data from the PLM acts as the carrier of knowledge, which is used to train a task-specific model with orders of magnitude fewer parameters than the PLM, achieving both higher performance and efficiency than prompt-based zero-shot learning methods on PLMs.
ZIN: When and How to Learn Invariance Without Environment Partition?
When data are divided into distinct environments according to the heterogeneity, recent invariant learning methods have proposed to learn robust and invariant models based on this environment partition.
Black-box Prompt Learning for Pre-trained Language Models
Particularly, instead of fine-tuning the model in the cloud, we adapt PLMs by prompt learning, which efficiently optimizes only a few parameters of the discrete prompts.
Bayesian Invariant Risk Minimization
Generalization under distributional shift is an open challenge for machine learning.
Homologies of path complexes and digraphs
In this paper we introduce a path complex that can be regarded as a generalization of the notion of a simplicial complex.
Combinatorics Algebraic Topology
