论文检索

Probabilistic Methods

Shengbo Wang, Jose Blanchet, Peter W Glynn

Overparameterized stochastic differential equation (SDE) models have achieved remarkable success in various complex environments, such as PDE-constrained optimization, stochastic control and reinforcement learning, financial engineering, and neural SDEs. These models often feature system evolution coefficients that are parameterized by a high-dimensional vector $\theta \in \mathbb{R}^n$, aiming to optimize expectations of the SDE, such as a value function, through stochastic gradient ascent. Consequently, designing efficient gradient estimators for which the computational complexity scales well with $n$ is of significant interest. This paper introduces a novel unbiased stochastic gradient estimator—the generator gradient estimator—for which the computation time remains stable in $n$. In addition to establishing the validity of our methodology for general SDEs with jumps, we also perform numerical experiments that test our estimator in linear-quadratic control problems parameterized by high-dimensional neural networks. The results show a significant improvement in efficiency compared to the widely used pathwise differentiation method: Our estimator achieves near-constant computation times, increasingly outperforms its counterpart as $n$ increases, and does so without compromising estimation variance. These empirical findings highlight the potential of our proposed methodology for optimizing SDEs in contemporary applications.

Theory · Domain Adaptation and Transfer Learning

Jin Shin, Hyun Kim

Test-time adaptation (TTA) is the most realistic methodology for adapting deep learning models to the real world using only unlabeled data from the target domain. Numerous TTA studies in deep learning have aimed at minimizing entropy. However, this necessitates forward/backward processes across the entire model and is limited by the incapability to fully leverage data based solely on entropy. This study presents a groundbreaking TTA solution that involves a departure from the conventional focus on minimizing entropy. Our innovative approach uniquely remodels the stem layer (i.e., the first layer) to emphasize minimizing a new learning criterion, namely, uncertainty. This method requires minimal involvement of the model's backbone, with only the stem layer participating in the TTA process. This approach significantly reduces the memory required for training and enables rapid adaptation to the target domain with minimal parameter updates. Moreover, to maximize data leveraging, the stem layer applies a discrete wavelet transform to the input features. It extracts multi-frequency domains and focuses on minimizing their individual uncertainties. The proposed method integrated into ResNet-26 and ResNet-50 models demonstrates its robustness by achieving outstanding TTA performance while using the least amount of memory compared to existing studies on CIFAR-10-C, ImageNet-C, and Cityscapes-C benchmark datasets. The code is available at https://github.com/janus103/L_TTA.

Language

Dan Zhang, Sining Zhoubian, Ziniu Hu, Yisong Yue, Yuxiao Dong, Jie Tang

Recent methodologies in LLM self-training mostly rely on LLM generating responses and filtering those with correct output answers as training data. This approach often yields a low-quality fine-tuning training set (e.g., incorrect plans or intermediate reasoning). In this paper, we develop a reinforced self-training approach, called ReST-MCTS*, based on integrating process reward guidance with tree search MCTS* for collecting higher-quality reasoning traces as well as per-step value to train policy and reward models. ReST-MCTS* circumvents the per-step manual annotation typically used to train process rewards by tree-search-based reinforcement learning: Given oracle final correct answers, ReST-MCTS* is able to infer the correct process rewards by estimating the probability this step can help lead to the correct answer. These inferred rewards serve dual purposes: they act as value targets for further refining the process reward model and also facilitate the selection of high-quality traces for policy model self-training. We first show that the tree-search policy in ReST-MCTS* achieves higher accuracy compared with prior LLM reasoning baselines such as Best-of-N and Tree-of-Thought, within the same search budget. We then show that by using traces searched by this tree-search policy as training data, we can continuously enhance the three language models for multiple iterations, and outperform other self-training algorithms such as ReST$^\text{EM}$ and Self-Rewarding LM.

Fairness

Cheng Li, Mengzhuo Chen, Jindong Wang, Sunayana Sitaram, Xing Xie

Large language models (LLMs) have been observed to exhibit bias towards certain cultures due to the predominance of training data obtained from English corpora. Considering that multilingual cultural data is often expensive to procure, existing methodologies address this challenge through prompt engineering or culture-specific pre-training. However, these strategies may neglect the knowledge deficiency of low-resource cultures and necessitate substantial computing resources. In this paper, we propose CultureLLM, a cost-effective solution to integrate cultural differences into LLMs. CultureLLM employs the World Value Survey (WVS) as seed data and generates semantically equivalent training data through the proposed semantic data augmentation. Utilizing only $50$ seed samples from WVS with augmented data, we fine-tune culture-specific LLMs as well as a unified model (CultureLLM-One) for $9$ cultures, encompassing both rich and low-resource languages. Extensive experiments conducted on $60$ culture-related datasets reveal that CultureLLM significantly surpasses various counterparts such as GPT-3.5 (by $8.1$\%) and Gemini Pro (by $9.5$\%), demonstrating performance comparable to or exceeding that of GPT-4. Our human study indicates that the generated samples maintain semantic equivalence to the original samples, offering an effective solution for LLMs augmentation. Code is released at https://github.com/Scarelette/CultureLLM.

Deep Learning · Algorithms

Usama Muneeb, Mesrob I Ohannessian

We consider scenarios where a very accurate (often small) predictive model using restricted features is available when training a full-featured (often larger) model. This restricted model may be thought of as ``side-information'', and can come either from an auxiliary dataset or from the same dataset by forcing the restriction. How can the restricted model be useful to the full model? To answer this, we introduce a methodology called Induced Model Matching (IMM). IMM aligns the context-restricted, or induced, version of the large model with the restricted model. We relate IMM to approaches such as noising, which is implicit in addressing the problem, and reverse knowledge distillation from weak teachers, which is explicit but does not exploit restriction being the nature of the weakness. We show that these prior methods can be thought of as approximations to IMM and can be problematic in terms of consistency. Experimentally, we first motivate IMM using logistic regression as a toy example. We then explore it in language modeling, the application that initially inspired it, and demonstrate it on both LSTM and transformer full models, using bigrams as restricted models. We lastly give a simple RL example, which shows that POMDP policies can help learn better MDP policies. The IMM principle is thus generally applicable in common scenarios where restricted data is cheaper to collect or restricted models are easier to learn.

Generative Models · Reasoning

Haoang Chi, He Li, Wenjing Yang, Feng Liu, Long Lan, Xiaoguang Ren, Tongliang Liu, Bo Han

Causal reasoning capability is critical in advancing large language models (LLMs) towards artificial general intelligence (AGI). While versatile LLMs appear to have demonstrated capabilities in understanding contextual causality and providing responses that obey the laws of causality, it remains unclear whether they perform genuine causal reasoning akin to humans. However, current evidence indicates the contrary. Specifically, LLMs are only capable of performing shallow (level-1) causal reasoning, primarily attributed to the causal knowledge embedded in their parameters, but they lack the capacity for genuine human-like (level-2) causal reasoning. To support this hypothesis, methodologically, we delve into the autoregression mechanism of transformer-based LLMs, revealing that it is not inherently causal. Empirically, we introduce a new causal Q&A benchmark named CausalProbe 2024, whose corpus is fresh and nearly unseen for the studied LLMs. Empirical results show a significant performance drop on CausalProbe 2024 compared to earlier benchmarks, indicating that LLMs primarily engage in level-1 causal reasoning.To bridge the gap towards level-2 causal reasoning, we draw inspiration from the fact that human reasoning is usually facilitated by general knowledge and intended goals. Inspired by this, we propose G$^2$-Reasoner, a LLM causal reasoning method that incorporates general knowledge and goal-oriented prompts into LLMs' causal reasoning processes. Experiments demonstrate that G$^2$-Reasoner significantly enhances LLMs' causal reasoning capability, particularly in fresh and fictitious contexts. This work sheds light on a new path for LLMs to advance towards genuine causal reasoning, going beyond level-1 and making strides towards level-2.

Applications

Michael Shalyt, Uri Seligmann, Itay Beit Halachmi, Ofir David, Rotem Elimelech, Ido Kaminer

Ongoing efforts that span over decades show a rise of AI methods for accelerating scientific discovery, yet accelerating discovery in mathematics remains a persistent challenge for AI.Specifically, AI methods were not effective in creation of formulas for mathematical constants because each such formula must be correct for infinite digits of precision, with 'near-true' formulas providing no insight toward the correct ones. Consequently, formula discovery lacks a clear distance metric needed to guide automated discovery in this realm.In this work, we propose a systematic methodology for categorization, characterization, and pattern identification of such formulas. The key to our methodology is introducing metrics based on the convergence dynamics of the formulas, rather than on the numerical value of the formula. These metrics enable the first automated clustering of mathematical formulas.We demonstrate this methodology on Polynomial Continued Fraction formulas, which are ubiquitous in their intrinsic connections to mathematical constants, and generalize many mathematical functions and structures.We test our methodology on a set of 1,768,900 such formulas, identifying many known formulas for mathematical constants, and discover previously unknown formulas for $\pi$, $\ln(2)$, Gauss', and Lemniscate's constants. The uncovered patterns enable a direct generalization of individual formulas to infinite families, unveiling rich mathematical structures. This success paves the way towards a generative model that creates formulas fulfilling specified mathematical properties, accelerating the rate of discovery of useful formulas.

Computer Vision · Segmentation

Kangning Liu, Brian Price, Jason Kuen, Yifei Fan, Zijun Wei, Luis Figueroa, Krzysztof Geras, Carlos Fernandez-Granda

The Segment Anything Model (SAM) is a large-scale foundation model that has revolutionized segmentation methodology. Despite its impressive generalization ability, the segmentation accuracy of SAM on images with intricate structures is often unsatisfactory. Recent works have proposed lightweight fine-tuning using high-quality annotated data to improve accuracy on such images. However, here we provide extensive empirical evidence that this strategy leads to forgetting how to "segment anything": these models lose the original generalization abilities of SAM, in the sense that they perform worse for segmentation tasks not represented in the annotated fine-tuning set. To improve performance without forgetting, we introduce a novel framework that combines high-quality annotated data with a large unlabeled dataset. The framework relies on two methodological innovations. First, we quantify the uncertainty in the SAM pseudo labels associated with the unlabeled data and leverage it to perform uncertainty-aware fine-tuning. Second, we encode the type of segmentation task associated with each training example using a $\textit{task prompt}$ to reduce ambiguity. We evaluated the proposed Segmentation with Uncertainty Model (SUM) on a diverse test set consisting of 14 public benchmarks, where it achieves state-of-the-art results. Notably, our method consistently surpasses SAM by 3-6 points in mean IoU and 4-7 in mean boundary IoU across point-prompt interactive segmentation rounds. Code is available at https://github.com/Kangningthu/SUM

Trustworthy Machine Learning

Hoin Jung, Taeuk Jang, Xiaoqian Wang

Recent advancements in Vision-Language Models (VLMs) have enabled complex multimodal tasks by processing text and image data simultaneously, significantly enhancing the field of artificial intelligence. However, these models often exhibit biases that can skew outputs towards societal stereotypes, thus necessitating debiasing strategies. Existing debiasing methods focus narrowly on specific modalities or tasks, and require extensive retraining. To address these limitations, this paper introduces Selective Feature Imputation for Debiasing (SFID), a novel methodology that integrates feature pruning and low confidence imputation (LCI) to effectively reduce biases in VLMs. SFID is versatile, maintaining the semantic integrity of outputs and costly effective by eliminating the need for retraining. Our experimental results demonstrate SFID's effectiveness across various VLMs tasks including zero-shot classification, text-to-image retrieval, image captioning, and text-to-image generation, by significantly reducing gender biases without compromising performance. This approach not only enhances the fairness of VLMs applications but also preserves their efficiency and utility across diverse scenarios.

Generative Models · New Approaches

Sapana Chaudhary, Ujwal Dinesha, Dileep Kalathil, Srinivas Shakkottai

We consider the challenge of mitigating the generation of negative or toxic content by the Large Language Models (LLMs) in response to certain prompts. We propose integrating risk-averse principles into LLM fine-tuning to minimize the occurrence of harmful outputs, particularly rare but significant events. By optimizing the risk measure of Conditional Value at Risk (CVaR), our methodology trains LLMs to exhibit superior performance in avoiding toxic outputs while maintaining effectiveness in generative tasks. Empirical evaluations on sentiment modification and toxicity mitigation tasks demonstrate the efficacy of risk-averse reinforcement learning with human feedback (RLHF) in promoting a safer and more constructive online discourse environment.

Deep Learning · Representation Learning

Bozhen Hu, Cheng Tan, Jun Xia, Yue Liu, Lirong Wu, Jiangbin Zheng, Yongjie Xu, Yufei Huang, Stan Z. Li

Learning effective representations is imperative for comprehending proteins and deciphering their biological functions. Recent strides in language models and graph neural networks have empowered protein models to harness primary or tertiary structure information for representation learning. Nevertheless, the absence of practical methodologies to appropriately model intricate inter-dependencies between protein sequences and structures has resulted in embeddings that exhibit low performance on tasks such as protein function prediction. In this study, we introduce CoupleNet, a novel framework designed to interlink protein sequences and structures to derive informative protein representations. CoupleNet integrates multiple levels and scales of features in proteins, encompassing residue identities and positions for sequences, as well as geometric representations for tertiary structures from both local and global perspectives. A two-type dynamic graph is constructed to capture adjacent and distant sequential features and structural geometries, achieving completeness at the amino acid and backbone levels. Additionally, convolutions are executed on nodes and edges simultaneously to generate comprehensive protein embeddings. Experimental results on benchmark datasets showcase that CoupleNet outperforms state-of-the-art methods, exhibiting particularly superior performance in low-sequence similarities scenarios, adeptly identifying infrequently encountered functions and effectively capturing remote homology relationships in proteins.

Generative Models · Reasoning

Xuefei Ning, Zifu Wang, Shiyao Li, Zinan Lin, Peiran Yao, Tianyu Fu, Matthew Blaschko, Guohao Dai, Huazhong Yang, Yu Wang

Teaching to improve student models (e.g., knowledge distillation) is an extensively studied methodology in LLMs. However, in human education, teaching enhances not only the students but also the teachers by fostering more rigorous and clearer reasoning, as well as deeper knowledge building. We ask: Can LLMs also learn by teaching (LbT) for better reasoning? If the answer is yes, we can potentially unlock the possibility of continuously advancing the models without solely relying on human-produced data or stronger models. In this paper, we provide a preliminary exploration of this question. We show that LbT ideas can be incorporated into existing LLM training/prompting pipelines and bring improvements. Specifically, we design three methods, each mimicking one of the three levels of LbT: observing students' feedback, learning from the feedback, and learning iteratively, with the goal of improving answer accuracy without training or improving models' inherent capability with fine-tuning. We reveal some findings: (1) Teaching materials that make it easier for students to learn (via in-context learning) have clearer and more accurate logic; (2) Weak-to-strong generalization: LbT might help improve strong models by teaching weak models; (3) Diversity in students might help: teaching multiple students could be better than teaching a single student or the teacher alone. We hope that our exploration can inspire future research on LbT and, more broadly, the adoption of advanced education techniques to improve LLMs. The code and website are at https://github.com/imagination-research/lbt and https://sites.google.com/view/llm-learning-by-teaching.

Language · Knowledge

Yuqing Yang, Ethan Chern, Xipeng Qiu, Graham Neubig, Pengfei Liu

Recent research has made significant strides in aligning large language models (LLMs) with helpfulness and harmlessness. In this paper, we argue for the importance of alignment for \emph{honesty}, ensuring that LLMs proactively refuse to answer questions when they lack knowledge, while still not being overly conservative. However, a pivotal aspect of alignment for honesty involves discerning an LLM's knowledge boundaries, which demands comprehensive solutions in terms of metric development, benchmark creation, and training methodologies. We address these challenges by first establishing a precise problem definition and defining ``honesty'' inspired by the Analects of Confucius. This serves as a cornerstone for developing metrics that effectively measure an LLM's honesty by quantifying its progress post-alignment. Furthermore, we introduce a flexible training framework which is further instantiated by several efficient fine-tuning techniques that emphasize honesty without sacrificing performance on other tasks. Our extensive experiments reveal that these aligned models show a marked increase in honesty, as indicated by our proposed metrics. We open-source all relevant resources to facilitate future research at \url{https://github.com/GAIR-NLP/alignment-for-honesty}.

Language · Generation

Xin Cheng, Xun Wang, Xingxing Zhang, Tao Ge, Si-Qing Chen, Furu Wei, Huishuai Zhang, Dongyan Zhao

This paper introduces xRAG, an innovative context compression method tailored for retrieval-augmented generation. xRAG reinterprets document embeddings in dense retrieval--traditionally used solely for retrieval--as features from the retrieval modality. By employing a modality fusion methodology, xRAG seamlessly integrates these embeddings into the language model representation space, effectively eliminating the need for their textual counterparts and achieving an extreme compression rate. In xRAG, the only trainable component is the modality bridge, while both the retriever and the language model remain frozen. This design choice allows for the reuse of offline-constructed document embeddings and preserves the plug-and-play nature of retrieval augmentation. Experimental results demonstrate that xRAG achieves an average improvement of over 10% across six knowledge-intensive tasks, adaptable to various language model backbones, ranging from a dense 7B model to an 8x7B Mixture of Experts configuration. xRAG not only significantly outperforms previous context compression methods but also matches the performance of uncompressed models on several datasets, while reducing overall FLOPs by a factor of 3.53. Our work pioneers new directions in retrieval-augmented generation from the perspective of multimodality fusion, and we hope it lays the foundation for future efficient and scalable retrieval-augmented systems.

Learning Theory

Andrew Bond, Zafer Dogan

Subspace learning is a critical endeavor in contemporary machine learning, particularly given the vast dimensions of modern datasets. In this study, we delve into the training dynamics of a single-layer GAN model from the perspective of subspace learning, framing these GANs as a novel approach to this fundamental task. Through a rigorous scaling limit analysis, we offer insights into the behavior of this model. Extending beyond prior research that primarily focused on sequential feature learning, we investigate the non-sequential scenario, emphasizing the pivotal role of inter-feature interactions in expediting training and enhancing performance, particularly with an uninformed initialization strategy. Our investigation encompasses both synthetic and real-world datasets, such as MNIST and Olivetti Faces, demonstrating the robustness and applicability of our findings to practical scenarios. By bridging our analysis to the realm of subspace learning, we systematically compare the efficacy of GAN-based methods against conventional approaches, both theoretically and empirically. Notably, our results unveil that while all methodologies successfully capture the underlying subspace, GANs exhibit a remarkable capability to acquire a more informative basis, owing to their intrinsic ability to generate new data samples. This elucidates the unique advantage of GAN-based approaches in subspace learning tasks.

Generative Models · In Context Learning

Wenyang Hu, Yao Shu, Zongmin Yu, Zhaoxuan Wu, Xiaoqiang Lin, Zhongxiang Dai, See-Kiong Ng, Bryan Kian Hsiang Low

The efficacy of large language models (LLMs) in understanding and generating natural language has aroused a wide interest in developing prompt-based methods to harness the power of black-box LLMs. Existing methodologies usually prioritize a global optimization for finding the global optimum, which however will perform poorly in certain tasks. This thus motivates us to re-think the necessity of finding a global optimum in prompt optimization. To answer this, we conduct a thorough empirical study on prompt optimization and draw two major insights. Contrasting with the rarity of global optimum, local optima are usually prevalent and well-performed, which can be more worthwhile for efficient prompt optimization (**Insight I**). The choice of the input domain, covering both the generation and the representation of prompts, affects the identification of well-performing local optima (**Insight II**). Inspired by these insights, we propose a novel algorithm, namely localized zeroth-order prompt optimization (ZOPO), which incorporates a Neural Tangent Kernel-based derived Gaussian process into standard zeroth-order optimization for an efficient search of well-performing local optima in prompt optimization. Remarkably, ZOPO outperforms existing baselines in terms of both the optimization performance and the query efficiency, which we demonstrate through extensive experiments.

Generative Models · Reasoning

Zhiqing Sun, Longhui Yu, Yikang Shen, Weiyang Liu, Yiming Yang, Sean Welleck, Chuang Gan

Current AI alignment methodologies rely on human-provided demonstrations or judgments, and the learned capabilities of AI systems would be upper-bounded by human capabilities as a result. This raises a challenging research question: How can we keep improving the systems when their capabilities have surpassed the levels of humans? This paper answers this question in the context of tackling hard reasoning tasks (e.g., level 4-5 MATH problems) via learning from human annotations on easier tasks (e.g., level 1-3 MATH problems), which we term as easy-to-hard generalization. Our key insight is that an evaluator (reward model) trained on supervisions for easier tasks can be effectively used for scoring candidate solutions of harder tasks and hence facilitating easy-to-hard generalization over different levels of tasks. Based on this insight, we propose a novel approach to scalable alignment, which firstly trains the (process-supervised) reward models on easy problems (e.g., level 1-3), and then uses them to evaluate the performance of policy models on hard problems. We show that such easy-to-hard generalization from evaluators can enable easy-to-hard generalizations in generators either through re-ranking or reinforcement learning (RL). Notably, our process-supervised 7b RL model and 34b model (reranking@1024) achieves an accuracy of 34.0% and 52.5% on MATH500, respectively, despite only using human supervision on easy problems. Our approach suggests a promising path toward AI systems that advance beyond the frontier of human supervision.

Optimization · Convex

Tianbo Li, Zekun Shi, Jiaxi Zhao, Min Lin

Performing eigendecomposition during neural network training is essential for tasks such as dimensionality reduction, network compression, image denoising, and graph learning. However, eigendecomposition is computationally expensive as it is orders of magnitude slower than other neural network operations. To address this challenge, we propose a novel approach called "amortized eigendecomposition" that relaxes the exact eigendecomposition by introducing an additional loss term called eigen loss. Our approach offers significant speed improvements by replacing the computationally expensive eigendecomposition with a more affordable QR decomposition at each iteration. Theoretical analysis guarantees that the desired eigenpair is attained as optima of the eigen loss. Empirical studies on nuclear norm regularization, latent-space principal component analysis, and graphs adversarial learning demonstrate significant improvements in training efficiency while producing nearly identical outcomes to conventional approaches. This novel methodology promises to integrate eigendecomposition efficiently into neural network training, overcoming existing computational challenges and unlocking new potential for advanced deep learning applications.

Language · Speech

Nicholas Babaev, Kirill Tamogashev, Azat Saginbaev, Ivan Shchekotov, Hanbin Bae, Hosang Sung, WonJun Lee, Hoon-Young Cho, Pavel Andreev

In this paper, we address the challenge of speech enhancement in real-world recordings, which often contain various forms of distortion, such as background noise, reverberation, and microphone artifacts.We revisit the use of Generative Adversarial Networks (GANs) for speech enhancement and theoretically show that GANs are naturally inclined to seek the point of maximum density within the conditional clean speech distribution, which, as we argue, is essential for speech enhancement task.We study various feature extractors for perceptual loss to facilitate the stability of adversarial training, developing a methodology for probing the structure of the feature space.This leads us to integrate WavLM-based perceptual loss into MS-STFT adversarial training pipeline, creating an effective and stable training procedure for the speech enhancement model.The resulting speech enhancement model, which we refer to as FINALLY, builds upon the HiFi++ architecture, augmented with a WavLM encoder and a novel training pipeline.Empirical results on various datasets confirm our model's ability to produce clear, high-quality speech at 48 kHz, achieving state-of-the-art performance in the field of speech enhancement. Demo page: https://samsunglabs.github.io/FINALLY-page/

Zirui Liu, Yan Zhuang, Qi Liu, Jiatong Li, Yuren Zhang, Zhenya Huang, Jinze Wu, Shijin Wang

As the deep integration of machine learning and intelligent education, Computerized Adaptive Testing (CAT) has received more and more research attention. Compared to traditional paper-and-pencil tests, CAT can deliver both personalized and interactive assessments by automatically adjusting testing questions according to the performance of students during the test process. Therefore, CAT has been recognized as an efficient testing methodology capable of accurately estimating a student’s ability with a minimal number of questions, leading to its widespread adoption in mainstream selective exams such as the GMAT and GRE. However, just improving the accuracy of ability estimation is far from satisfactory in the real-world scenarios, since an accurate ranking of students is usually more important (e.g., in high-stakes exams). Considering the shortage of existing CAT solutions in student ranking, this paper emphasizes the importance of aligning test outcomes (student ranks) with the true underlying abilities of students. Along this line, different from the conventional independent testing paradigm among students, we propose a novel collaborative framework, Collaborative Computerized Adaptive Testing (CCAT), that leverages inter-student information to enhance student ranking. By using collaborative students as anchors to assist in ranking test-takers, CCAT can give both theoretical guarantees and experimental validation for ensuring ranking consistency.