相关论文: The Labyrinth and the Thread: Rethinking Regulariz…

Sequential learning, also called lifelong learning, studies the problem of learning tasks in a sequence with access restricted to only the data of the current task. In this paper we look at a scenario with fixed model capacity, and…

Understanding how ideas develop and flow in small-group conversations is critical for analyzing collaborative learning. A key structural feature of these interactions is threading, the way discourse talk naturally organizes into interwoven…

We propose a novel data-dependent structured gradient regularizer to increase the robustness of neural networks vis-a-vis adversarial perturbations. Our regularizer can be derived as a controlled approximation from first principles,…

Given the impressive capabilities of recent Large Language Models (LLMs), we investigate and benchmark the most popular proprietary and different sized open source models on the task of explicit instruction following in conflicting…

Unstructured Knowledge Editing (UKE) is crucial for updating the relevant knowledge of large language models (LLMs). It focuses on unstructured inputs, such as long or free-form texts, which are common forms of real-world knowledge.…

Given the prevalence of crowd sourced labor in creating Natural Language processing datasets, these aforementioned sets have become increasingly large. For instance, the SQUAD dataset currently sits at over 80,000 records. However, because…

Large language models (LLMs) are a promising venue for natural language understanding and generation tasks. However, current LLMs are far from reliable: they are prone to generate non-factual information and, more crucially, to contradict…

Temporal difference (TD) learning is a foundational algorithm in reinforcement learning (RL). For nearly forty years, TD learning has served as a workhorse for applied RL as well as a building block for more complex and specialized…

Robust fine-tuning aims to achieve competitive in-distribution (ID) performance while maintaining the out-of-distribution (OOD) robustness of a pre-trained model when transferring it to a downstream task. To remedy this, most robust…

When optimizing a thread in a concurrent program (either done manually or by the compiler), it must be guaranteed that the resulting thread is a refinement of the original thread. Most theories of valid optimizations are formulated in terms…

Real-world sequential signals, such as audio or video, contain critical information that is often embedded within long periods of silence or noise. While recurrent neural networks (RNNs) are designed to process such data efficiently, they…

Reasoning has become a central capability in large language models. Recent research has shown that reasoning performance can be improved by looping an LLM's layers in the latent dimension, resulting in looped reasoning language models.…

With the increasing capabilities of Large Language Models (LLMs), parallel reasoning has emerged as a new inference paradigm that enhances reasoning robustness by concurrently exploring multiple lines of thought before converging on a final…

Recent advances in artificial intelligence (AI) and natural language processing (NLP) have enabled tools to support systematic literature reviews (SLRs), yet existing frameworks often produce outputs that are efficient but contextually…

Large language models (LLMs) are increasingly being used to synthesize and reason about source code. However, the static nature of these models' knowledge does not reflect the fact that libraries and API functions they invoke are…

Text-based editing diffusion models exhibit limited performance when the user's input instruction is ambiguous. To solve this problem, we propose $\textit{Specify ANd Edit}$ (SANE), a zero-shot inference pipeline for diffusion-based editing…

A sequence $e_0,e_1,\ldots$ of edit operations in a string $T$ is called ephemeral if operation $e_i$ constructing string $T^i$, for all $i=2k$ with $k\in\mathbb{N}$, is reverted by operation $e_{i+1}$ that reconstructs $T$. Such a sequence…

Large language models have demonstrated impressive reasoning capabilities but are inherently limited by their knowledge reservoir. Retrieval-augmented reasoning mitigates this limitation by allowing LLMs to query external resources, but…

The organization of latent knowledge within large-scale models poses unique challenges when addressing overlapping representations and optimizing contextual accuracy. Conceptual redundancies embedded across layers often result in…

Reasoning over procedural sequences, where the order of steps directly impacts outcomes, is a critical capability for large language models (LLMs). In this work, we study the task of reconstructing globally ordered sequences from shuffled…