arXiv subjects
Publications and source records attributed to Zheng Liu.
We construct four-variable $p$-adic $L$-functions for cuspidal Hida families on ${\rm GSp}(4)\times{\rm GL}(2)$ and prove a complete interpolation formula. The archimedean zeta integrals are computed by using a partial interpolation formula for the four-variable $p$-adic $L$-functions and some previously constructed $p$-adic $L$-functions.
Multi-modal retrieval becomes increasingly popular in practice. However, the existing retrievers are mostly text-oriented, which lack the capability to process visual information. Despite the presence of vision-language models like CLIP, the current methods are severely limited in representing the text-only and image-only data. In this work, we present a new embedding model VISTA for universal multi-modal retrieval. Our work brings forth threefold technical contributions. Firstly, we introduce a flexible architecture which extends a powerful text encoder with the image understanding capability by introducing visual token embeddings. Secondly, we develop two data generation strategies, which bring high-quality composed image-text to facilitate the training of the embedding model. Thirdly, we introduce a multi-stage training algorithm, which first aligns the visual token embedding with the text encoder using massive weakly labeled data, and then develops multi-modal representation capability using the generated composed image-text data. In our experiments, VISTA achieves superior performances across a variety of multi-modal retrieval tasks in both zero-shot and supervised settings. Our model, data, and source code are available at https://github.com/FlagOpen/FlagEmbedding.
The exceptional properties of two-dimensional (2D) magnet materials present a novel approach to fabricate functional magnetic tunnel junctions (MTJ) by constructing full van der Waals (vdW) heterostructures with atomically sharp and clean interfaces. The exploration of vdW MTJ devices with high working temperature and adjustable functionalities holds great potential for advancing the application of 2D materials in magnetic sensing and data storage. Here, we report the observation of highly tunable room-temperature tunneling magnetoresistance through electronic means in a full vdW Fe3GaTe2/WSe2/Fe3GaTe2 MTJ. The spin valve effect of the MTJ can be detected even with the current below 1 nA, both at low and room temperatures, yielding a tunneling magnetoresistance (TMR) of 340% at 2 K and 50% at 300 K, respectively. Importantly, the magnitude and sign of TMR can be modulated by a DC bias current, even at room temperature, a capability that was previously unrealized in full vdW MTJs. This tunable TMR arises from the contribution of energy-dependent localized spin states in the metallic ferromagnet Fe3GaTe2 during tunnel transport when a finite electrical bias is applied. Our work offers a new perspective for designing and exploring room-temperature tunable spintronic devices based on vdW magnet heterostructures.
Large language models (LLMs) have demonstrated impressive capabilities in various natural language processing tasks. Despite this, their application to information retrieval (IR) tasks is still challenging due to the infrequent occurrence of many IR-specific concepts in natural language. While prompt-based methods can provide task descriptions to LLMs, they often fall short in facilitating a comprehensive understanding and execution of IR tasks, thereby limiting LLMs' applicability. To address this gap, in this work, we explore the potential of instruction tuning to enhance LLMs' proficiency in IR tasks. We introduce a novel instruction tuning dataset, INTERS, encompassing 20 tasks across three fundamental IR categories: query understanding, document understanding, and query-document relationship understanding. The data are derived from 43 distinct datasets with manually written templates. Our empirical results reveal that INTERS significantly boosts the performance of various publicly available LLMs, such as LLaMA, Mistral, and Phi, in IR tasks. Furthermore, we conduct extensive experiments to analyze the effects of instruction design, template diversity, few-shot demonstrations, and the volume of instructions on performance. We make our dataset and the fine-tuned models publicly accessible at https://github.com/DaoD/INTERS.
The learning and deployment of long-LLMs remains a challenging problem despite recent progresses. In this work, we argue that the long-LLMs are not a necessity to solve long-context tasks, as common long-context tasks are short-context solvable, i.e. they can be solved by purely working with oracle short-contexts within the long-context tasks' inputs. On top of this argument, we propose a framework called LC-Boost (Long-Context Bootstrapper), which enables a short-LLM to address the long-context tasks in a bootstrapping manner. In our framework, the short-LLM prompts itself to reason for two critical decisions: 1) how to access to the appropriate part of context within the input, 2) how to make effective use of the accessed context. By adaptively accessing and utilizing the context based on the presented tasks, LC-Boost can serve as a general framework to handle diversified long-context processing problems. We comprehensively evaluate different types of tasks from popular long-context benchmarks, where LC-Boost is able to achieve a substantially improved performance with a much smaller consumption of resource.
We derive precise formulas for the archimedean Euler factors occurring in certain standard Langlands $L$-functions for unitary groups. In the 1980s, Paul Garrett, as well as Ilya Piatetski-Shapiro and Stephen Rallis (independently of Garrett), discovered integral representations of automorphic $L$-functions that are Eulerian but, in contrast to the Rankin--Selberg and Langlands--Shahidi methods, do not require that the automorphic representations to which the $L$-functions are associated are globally generic. Their approach, the doubling method, opened the door to a variety of applications that could not be handled by prior methods. For over three decades, though, the integrals occurring in the Euler factors at archimedean places for unitary groups eluded precise computation, except under particular simplifications (such as requiring certain representations to be one-dimensional, as Garrett did in the first major progress on this computation and only prior progress for general signatures). We compute these integrals for holomorphic discrete series of general vector weights for unitary groups of any signature. This has consequences not only for special values of $L$-functions in the archimedean setting, but also for $p$-adic $L$-functions, where the corresponding term had remained open.
Recent breakthroughs in large models have highlighted the critical significance of data scale, labels and modals. In this paper, we introduce MS MARCO Web Search, the first large-scale information-rich web dataset, featuring millions of real clicked query-document labels. This dataset closely mimics real-world web document and query distribution, provides rich information for various kinds of downstream tasks and encourages research in various areas, such as generic end-to-end neural indexer models, generic embedding models, and next generation information access system with large language models. MS MARCO Web Search offers a retrieval benchmark with three web retrieval challenge tasks that demand innovations in both machine learning and information retrieval system research domains. As the first dataset that meets large, real and rich data requirements, MS MARCO Web Search paves the way for future advancements in AI and system research. MS MARCO Web Search dataset is available at: https://github.com/microsoft/MS-MARCO-Web-Search.
We extend the context length of Llama-3-8B-Instruct from 8K to 80K via QLoRA fine-tuning. The entire training cycle is super efficient, which takes 8 hours on one 8xA800 (80G) GPU machine. The resulted model exhibits superior performances across a broad range of evaluation tasks, such as NIHS, topic retrieval, and long-context language understanding; meanwhile, it also well preserves the original capability over short contexts. The dramatic context extension is mainly attributed to merely 3.5K synthetic training samples generated by GPT-4 , which indicates the LLMs' inherent (yet largely underestimated) potential to extend its original context length. In fact, the context length could be extended far beyond 80K with more computation resources. Therefore, the team will publicly release the entire resources (including data, model, data generation pipeline, training code) so as to facilitate the future research from the community: \url{https://github.com/FlagOpen/FlagEmbedding}.
Large language models (LLMs) show early signs of artificial general intelligence but struggle with hallucinations. One promising solution to mitigate these hallucinations is to store external knowledge as embeddings, aiding LLMs in retrieval-augmented generation. However, such a solution risks compromising privacy, as recent studies experimentally showed that the original text can be partially reconstructed from text embeddings by pre-trained language models. The significant advantage of LLMs over traditional pre-trained models may exacerbate these concerns. To this end, we investigate the effectiveness of reconstructing original knowledge and predicting entity attributes from these embeddings when LLMs are employed. Empirical findings indicate that LLMs significantly improve the accuracy of two evaluated tasks over those from pre-trained models, regardless of whether the texts are in-distribution or out-of-distribution. This underscores a heightened potential for LLMs to jeopardize user privacy, highlighting the negative consequences of their widespread use. We further discuss preliminary strategies to mitigate this risk.
Conversational search requires accurate interpretation of user intent from complex multi-turn contexts. This paper presents ChatRetriever, which inherits the strong generalization capability of large language models to robustly represent complex conversational sessions for dense retrieval. To achieve this, we propose a simple and effective dual-learning approach that adapts LLM for retrieval via contrastive learning while enhancing the complex session understanding through masked instruction tuning on high-quality conversational instruction tuning data. Extensive experiments on five conversational search benchmarks demonstrate that ChatRetriever substantially outperforms existing conversational dense retrievers, achieving state-of-the-art performance on par with LLM-based rewriting approaches. Furthermore, ChatRetriever exhibits superior robustness in handling diverse conversational contexts. Our work highlights the potential of adapting LLMs for retrieval with complex inputs like conversational search sessions and proposes an effective approach to advance this research direction.
Recently, generative retrieval emerges as a promising alternative to traditional retrieval paradigms. It assigns each document a unique identifier, known as DocID, and employs a generative model to directly generate the relevant DocID for the input query. A common choice for DocID is one or several natural language sequences, e.g. the title or n-grams, so that the pre-trained knowledge of the generative model can be utilized. However, a sequence is generated token by token, where only the most likely candidates are kept and the rest are pruned at each decoding step, thus, retrieval fails if any token within the relevant DocID is falsely pruned. What's worse, during decoding, the model can only perceive preceding tokens in DocID while being blind to subsequent ones, hence is prone to make such errors. To address this problem, we present a novel framework for generative retrieval, dubbed Term-Set Generation (TSGen). Instead of sequences, we use a set of terms as DocID, which are automatically selected to concisely summarize the document's semantics and distinguish it from others. On top of the term-set DocID, we propose a permutation-invariant decoding algorithm, with which the term set can be generated in any permutation yet will always lead to the corresponding document. Remarkably, TSGen perceives all valid terms rather than only the preceding ones at each decoding step. Given the constant decoding space, it can make more reliable decisions due to the broader perspective. TSGen is also resilient to errors: the relevant DocID will not be pruned as long as the decoded term belongs to it. Lastly, we design an iterative optimization procedure to incentivize the model to generate the relevant term set in its favorable permutation. We conduct extensive experiments on popular benchmarks, which validate the effectiveness, the generalizability, the scalability, and the efficiency of TSGen.
Two-dimensional electron systems (2DES) are promising for investigating correlated quantum phenomena. In particular, 2D oxides provide a platform that can host various quantum phases such as quantized Hall effect, superconductivity, or magnetism. The realization of such quantum phases in 2D oxides heavily relies on dedicated heterostructure growths. Here we show the integer quantum Hall effect achieved in chemical vapor deposition grown Bi2O2Se - a representative member of a more accessible oxide family. A single or few sub-band 2DES can be prepared in thin films of Bi2O2Se, where the film thickness acts as the sole design parameter and the sub-band occupation is determined by the electric field effect. This new oxide platform exhibits characteristic advantages in structural flexibility due to its layered nature, making it suitable for scalable growth. The unique small mass distinguishes Bi2O2Se from other high-mobility oxides, providing a new platform for exploring quantum Hall physics in 2D oxides.
The intimate connection between magnetism and superconducting pairing routinely plays a central role in determining the occurrence of unconventional superconducting states. In high-transition-temperature (high-Tc) stripe-ordered cuprate superconductors and a magnetically ordered iron-based superconductor, the coupling between magnetism and superconductivity gives birth to novel phases of matter with modulation of the superconducting pairing in the real space. Further exploration of these phases can shed light on the mechanism of unconventional superconductivity. Here we report on the discovery of a peculiar spatially-varying superconducting state residing at the interface between (110)-oriented KTaO3 and ferromagnetic EuO. Electrical transport measurements reveal different Tc and upper critical fields (Hc2) with current applied along the two orthogonal in-plane directions. Such anisotropy persistently occurs in the low-carrier-density samples that are characterized by strong coupling between Ta 5d and Eu 4f electrons, whereas in the high-carrier-density samples the coupling is weakened and Tc and Hc2 becomes isotropic. Complemented by local imaging of diamagnetism and theoretical analysis, our observations imply an unprecedented emergence of superconducting stripes wherein the phase coherence is established ahead of the rest of the interface, arising from a band-filling-dependent ferromagnetic proximity. The realization of such exotic superconducting states provides impetus for the study of novel physics in heterostructures possessing both magnetism and superconductivity.
The Anomalous Hall Effect (AHE) manifests as a transverse voltage proportional to magnetization in ferromagnetic materials under the application of a charge current, being an indispensable tool for probing magnetism, especially in nanoscale devices. However, the AHE primarily sensitizes to out-of-plane magnetization, thereby hindering its capacity to discern the in-plane magnetization, a characteristic prevalent in ferromagnetic films. Here we challenge this conventional understanding by demonstrating the in-plane magnetization-induced AHE in iron and nickel, two ubiquitous ferromagnets. This observation of the in-plane AHE is remarkable as it contradicts existing theories that forbid such phenomena in cubic crystal systems. We trace the origin of this unanticipated phenomenon to a hitherto unconsidered octupole of the anomalous Hall conductivity in the magnetization space, a mechanism we propose could enable the detection of in-plane AHE in a wide range of ferromagnetic materials. This work realizes the in-plane AHE in common ferromagnets by exploiting the anomalous Hall conductivity octupole, revealing a new physical origin of the AHE and promising to revolutionize the design of magnetic devices and sensors.
Large language models (LLMs) call for extension of context to handle many critical applications. However, the existing approaches are prone to expensive costs and inferior quality of context extension. In this work, we proposeExtensible Embedding, which realizes high-quality extension of LLM's context with strong flexibility and cost-effectiveness. Extensible embedding stand as an enhancement of typical token embedding, which represents the information for an extensible scope of context instead of a single token. By leveraging such compact input units of higher information density, the LLM can access to a vast scope of context even with a small context window. Extensible embedding is systematically optimized in architecture and training method, which leads to multiple advantages. 1) High flexibility of context extension, which flexibly supports ad-hoc extension of diverse context lengths. 2) Strong sample efficiency of training, which enables the embedding model to be learned in a cost-effective way. 3) Superior compatibility with the existing LLMs, where the extensible embedding can be seamlessly introduced as a plug-in component. Comprehensive evaluations on long-context language modeling and understanding tasks verify extensible embedding as an effective, efficient, flexible, and compatible method to extend the LLM's context.
Large language models (LLMs) call for extension of context to handle many critical applications. However, the existing approaches are prone to expensive costs and inferior quality of context extension. In this work, we propose Extensible Embedding, which realizes high-quality extension of LLM's context with strong flexibility and cost-effectiveness. Extensible embedding stand as an enhancement of typical token embedding, which represents the information for an extensible scope of context instead of a single token. By leveraging such compact input units of higher information density, the LLM can access to a vast scope of context even with a small context window. Extensible embedding is systematically optimized in architecture and training method, which leads to multiple advantages. 1) High flexibility of context extension, which flexibly supports ad-hoc extension of diverse context lengths. 2) Strong sample efficiency of training, which enables the embedding model to be learned in a cost-effective way. 3) Superior compatibility with the existing LLMs, where the extensible embedding can be seamlessly introduced as a plug-in component. Comprehensive evaluations on long-context language modeling and understanding tasks verify extensible embedding as an effective, efficient, flexible, and compatible method to extend the LLM's context.
Retrieval-augmented generation have become central in natural language processing due to their efficacy in generating factual content. While traditional methods employ single-time retrieval, more recent approaches have shifted towards multi-time retrieval for multi-hop reasoning tasks. However, these strategies are bound by predefined reasoning steps, potentially leading to inaccuracies in response generation. This paper introduces MetaRAG, an approach that combines the retrieval-augmented generation process with metacognition. Drawing from cognitive psychology, metacognition allows an entity to self-reflect and critically evaluate its cognitive processes. By integrating this, MetaRAG enables the model to monitor, evaluate, and plan its response strategies, enhancing its introspective reasoning abilities. Through a three-step metacognitive regulation pipeline, the model can identify inadequacies in initial cognitive responses and fixes them. Empirical evaluations show that MetaRAG significantly outperforms existing methods.
This paper presents a novel Chunking-Free In-Context (CFIC) retrieval approach, specifically tailored for Retrieval-Augmented Generation (RAG) systems. Traditional RAG systems often struggle with grounding responses using precise evidence text due to the challenges of processing lengthy documents and filtering out irrelevant content. Commonly employed solutions, such as document chunking and adapting language models to handle longer contexts, have their limitations. These methods either disrupt the semantic coherence of the text or fail to effectively address the issues of noise and inaccuracy in evidence retrieval. CFIC addresses these challenges by circumventing the conventional chunking process. It utilizes the encoded hidden states of documents for in-context retrieval, employing auto-aggressive decoding to accurately identify the specific evidence text required for user queries, eliminating the need for chunking. CFIC is further enhanced by incorporating two decoding strategies, namely Constrained Sentence Prefix Decoding and Skip Decoding. These strategies not only improve the efficiency of the retrieval process but also ensure that the fidelity of the generated grounding text evidence is maintained. Our evaluations of CFIC on a range of open QA datasets demonstrate its superiority in retrieving relevant and accurate evidence, offering a significant improvement over traditional methods. By doing away with the need for document chunking, CFIC presents a more streamlined, effective, and efficient retrieval solution, making it a valuable advancement in the field of RAG systems.