016 namely CLEVER, which is augmentation-free 017 and mitigates biases on the inference stage. 018 Specifically, we train a claim-evidence fusion 019 model and a claim-only model …
Measuring Mathematical Problem Solving With the MATH Dataset
Oct 18, 2021 · To find the limits of Transformers, we collected 12,500 math problems. While a three-time IMO gold medalist got 90%, GPT-3 models got ~5%, with accuracy increasing slowly.
Alias-Free Mamba Neural Operator - OpenReview
Sep 25, 2024 · Functionally, MambaNO achieves a clever balance between global integration, facilitated by state space model of Mamba that scans the entire function, and local integration, …
LLaVA-OneVision: Easy Visual Task Transfer | OpenReview
Feb 9, 2025 · We present LLaVA-OneVision, a family of open large multimodal models (LMMs) developed by consolidating our insights into data, models, and visual representations in the …
Self-Taught Evaluators - OpenReview
Sep 26, 2024 · Model-based evaluation is at the heart of successful model development -- as a reward model for training, and as a replacement for human evaluation. To train such …
Let's reward step by step: Step-Level reward model as the...
Sep 24, 2023 · Recent years have seen considerable advancements in multi-step reasoning by Large Language Models (LLMs). Numerous studies elucidate the merits of integrating …
A Theoretically-Principled Sparse, Connected, and Rigid Graph ...
Jan 22, 2025 · Graph neural networks (GNNs) -- learn graph representations by exploiting the graph's sparsity, connectivity, and symmetries -- have become indispensable for learning …
Training Large Language Model to Reason in a Continuous
Sep 26, 2024 · Large language models are restricted to reason in the “language space”, where they typically express the reasoning process with a chain-of-thoughts (CoT) to solve a …
DEBERTA: DECODING-ENHANCED BERT WITH DISENTANGLED …
Jan 12, 2021 · Recent progress in pre-trained neural language models has significantly improved the performance of many natural language processing (NLP) tasks. In this paper we propose a …
Eureka: Human-Level Reward Design via Coding Large Language …
Jan 16, 2024 · Large Language Models (LLMs) have excelled as high-level semantic planners for sequential decision-making tasks. However, harnessing them to learn complex low-level …