M2K-VDG: Model-Adaptive Multimodal Knowledge Anchor Enhanced Video-grounded Dialogue Generation

Video-grounded dialogue generation (VDG) requires the system to generate a fluent and accurate answer based on multimodal knowledge. However, the difficulty in multimodal knowledge utilization brings serious hallucinations to VDG models in practice. Although previous works mitigate the hallucination in a variety of ways, they hardly take notice of the importance of the multimodal knowledge anchor answer tokens. In this paper, we reveal via perplexity that different VDG models experience varying hallucinations and exhibit diverse anchor tokens. Based on this observation, we propose M2K-VDG, a model-adaptive multimodal knowledge anchor enhancement framework for hallucination reduction. Furthermore, we introduce the counterfactual effect for more accurate anchor token detection. The experimental results on three popular benchmarks exhibit the superiority of our approach over state-of-the-art methods, demonstrating its effectiveness in reducing hallucinations.

DictLLM: Harnessing Key-Value Data Structures with Large Language Models for Enhanced Medical Diagnostics

Structured data offers a sophisticated mechanism for the organization of information. Existing methodologies for the text-serialization of structured data in the context of large language models fail to adequately address the heterogeneity inherent in key-value structured data. These methods are not ideal and frequently result in larger input sizes and poor adaptability to input changes. In this paper, we introduce DictLLM, an innovative framework designed to improve the modeling of key-value structured data, like medical laboratory reports, for generating medical diagnoses. DictLLM integrates three key components: (1) group positional encoding to maintain permutation invariance, (2) hierarchical attention bias to capture the inherent bias in structured data, and (3) an optimal transport alignment layer that aligns the embedding generated by the dictionary encoder with the LLM, thereby producing a sequence of fixed-length virtual tokens. We carry out experiments using various LLM models on a comprehensive real-world medical laboratory report dataset for automatic diagnosis generation, our findings illustrate that DictLLM significantly outperforms established baseline methods and few-shot GPT-4 implementations in terms of both Rouge-L and Knowledge F1 scores. Furthermore, our evaluation of the framework's scalability and robustness, through a series of experiments, underscores its exceptional capability in accurately modeling the complex key-value data structure of medical dictionary data.

Knowledge Graph-based Session Recommendation with Adaptive Propagation

Session-based recommender systems (SBRSs) predict users' next interacted items based on their historical activities. While most SBRSs capture purchasing intentions locally within each session, capturing items' global information across different sessions is crucial in characterizing their general properties. Previous works capture this cross-session information by constructing graphs and incorporating neighbor information. However, this incorporation cannot vary adaptively according to the unique intention of each session, and the constructed graphs consist of only one type of user-item interaction. To address these limitations, we propose knowledge graph-based session recommendation with session-adaptive propagation. Specifically, we build a knowledge graph by connecting items with multi-typed edges to characterize various user-item interactions. Then, we adaptively aggregate items' neighbor information considering user intention within the learned session. Experimental results demonstrate that equipping our constructed knowledge graph and session-adaptive propagation enhances session recommendation backbones by 10%-20%. Moreover, we provide an industrial case study showing our proposed framework achieves 2% performance boost over an existing well-deployed model at The Home Depot e-platform.

Rec-GPT4V: Multimodal Recommendation with Large Vision-Language Models

The development of large vision-language models (LVLMs) offers the potential to address challenges faced by traditional multimodal recommendations thanks to their proficient understanding of static images and textual dynamics. However, the application of LVLMs in this field is still limited due to the following complexities: First, LVLMs lack user preference knowledge as they are trained from vast general datasets. Second, LVLMs suffer setbacks in addressing multiple image dynamics in scenarios involving discrete, noisy, and redundant image sequences. To overcome these issues, we propose the novel reasoning scheme named Rec-GPT4V: Visual-Summary Thought (VST) of leveraging large vision-language models for multimodal recommendation. We utilize user history as in-context user preferences to address the first challenge. Next, we prompt LVLMs to generate item image summaries and utilize image comprehension in natural language space combined with item titles to query the user preferences over candidate items. We conduct comprehensive experiments across four datasets with three LVLMs: GPT4-V, LLaVa-7b, and LLaVa-13b. The numerical results indicate the efficacy of VST.

MEMORYLLM: Towards Self-Updatable Large Language Models

Existing Large Language Models (LLMs) usually remain static after deployment, which might make it hard to inject new knowledge into the model. We aim to build models containing a considerable portion of self-updatable parameters, enabling the model to integrate new knowledge effectively and efficiently. To this end, we introduce MEMORYLLM, a model that comprises a transformer and a fixed-size memory pool within the latent space of the transformer. MEMORYLLM can self-update with text knowledge and memorize the knowledge injected earlier. Our evaluations demonstrate the ability of MEMORYLLM to effectively incorporate new knowledge, as evidenced by its performance on model editing benchmarks. Meanwhile, the model exhibits long-term information retention capacity, which is validated through our custom-designed evaluations and long-context benchmarks. MEMORYLLM also shows operational integrity without any sign of performance degradation even after nearly a million memory updates.

LV-Eval: A Balanced Long-Context Benchmark with 5 Length Levels Up to 256K

State-of-the-art large language models (LLMs) are now claiming remarkable supported context lengths of 256k or even more. In contrast, the average context lengths of mainstream benchmarks are insufficient (5k-21k), and they suffer from potential knowledge leakage and inaccurate metrics, resulting in biased evaluation. This paper introduces LV-Eval, a challenging long-context benchmark with five length levels (16k, 32k, 64k, 128k, and 256k) reaching up to 256k words. LV-Eval features two main tasks, single-hop QA and multi-hop QA, comprising 11 bilingual datasets. The design of LV-Eval has incorporated three key techniques, namely confusing facts insertion, keyword and phrase replacement, and keyword-recall-based metric design. The advantages of LV-Eval include controllable evaluation across different context lengths, challenging test instances with confusing facts, mitigated knowledge leakage, and more objective evaluations. We evaluate 10 LLMs on LV-Eval and conduct ablation studies on the techniques used in LV-Eval construction. The results reveal that: (i) Commercial LLMs generally outperform open-source LLMs when evaluated within length levels shorter than their claimed context length. However, their overall performance is surpassed by open-source LLMs with longer context lengths. (ii) Extremely long-context LLMs, such as Yi-6B-200k, exhibit a relatively gentle degradation of performance, but their absolute performances may not necessarily be higher than those of LLMs with shorter context lengths. (iii) LLMs' performances can significantly degrade in the presence of confusing information, especially in the pressure test of "needle in a haystack". (iv) Issues related to knowledge leakage and inaccurate metrics introduce bias in evaluation, and these concerns are alleviated in LV-Eval. All datasets and evaluation codes are released at: https://github.com/infinigence/LVEval.

Cross-Domain Few-Shot Object Detection via Enhanced Open-Set Object Detector

This paper addresses the challenge of cross-domain few-shot object detection (CD-FSOD), aiming to develop an accurate object detector for novel domains with minimal labeled examples. While transformer-based open-set detectors e.g., DE-ViT~\cite{zhang2023detect} have excelled in both open-vocabulary object detection and traditional few-shot object detection, detecting categories beyond those seen during training, we thus naturally raise two key questions: 1) can such open-set detection methods easily generalize to CD-FSOD? 2) If no, how to enhance the results of open-set methods when faced with significant domain gaps? To address the first question, we introduce several metrics to quantify domain variances and establish a new CD-FSOD benchmark with diverse domain metric values. Some State-Of-The-Art (SOTA) open-set object detection methods are evaluated on this benchmark, with evident performance degradation observed across out-of-domain datasets. This indicates the failure of adopting open-set detectors directly for CD-FSOD. Sequentially, to overcome the performance degradation issue and also to answer the second proposed question, we endeavor to enhance the vanilla DE-ViT. With several novel components including finetuning, a learnable prototype module, and a lightweight attention module, we present an improved Cross-Domain Vision Transformer for CD-FSOD (CD-ViTO). Experiments show that our CD-ViTO achieves impressive results on both out-of-domain and in-domain target datasets, establishing new SOTAs for both CD-FSOD and FSOD. All the datasets, codes, and models will be released to the community.

Peer-review-in-LLMs: Automatic Evaluation Method for LLMs in Open-environment

Existing large language models (LLMs) evaluation methods typically focus on testing the performance on some closed-environment and domain-specific benchmarks with human annotations. In this paper, we explore a novel unsupervised evaluation direction, utilizing peer-review mechanisms to measure LLMs automatically. In this setting, both open-source and closed-source LLMs lie in the same environment, capable of answering unlabeled questions and evaluating each other, where each LLM's response score is jointly determined by other anonymous ones. To obtain the ability hierarchy among these models, we assign each LLM a learnable capability parameter to adjust the final ranking. We formalize it as a constrained optimization problem, intending to maximize the consistency of each LLM's capabilities and scores. The key assumption behind is that high-level LLM can evaluate others' answers more accurately than low-level ones, while higher-level LLM can also achieve higher response scores. Moreover, we propose three metrics called PEN, CIN, and LIS to evaluate the gap in aligning human rankings. We perform experiments on multiple datasets with these metrics, validating the effectiveness of the proposed approach.

Deep learning and random light structuring ensure robust free-space communications

Having shown early promise, free-space optical communications (FSO) face formidable challenges in the age of information explosion. The ever-growing demand for greater channel communication capacity is one of the challenges. The inter-channel crosstalk, which severely degrades the quality of transmitted information, creates another roadblock in the way of efficient FSO implementation. Here we advance theoretically and realize experimentally a potentially high-capacity FSO protocol that enables high-fidelity transfer of an image, or set of images through a complex environment. In our protocol, we complement random light structuring at the transmitter with a deep learning image classification platform at the receiver. Multiplexing novel, independent, mutually orthogonal degrees of freedom available to structured random light can potentially significantly boost the channel communication capacity of our protocol without introducing any deleterious crosstalk. Specifically, we show how one can multiplex the degrees of freedom associated with the source coherence radius and a spatial position of a beamlet within an array of structured random beams to greatly enhance the capacity of our communication link. The superb resilience of structured random light to environmental noise, as well as extreme efficiency of deep learning networks at classifying images guarantees high-fidelity image transfer within the framework of our protocol.

Disentangled Condensation for Large-scale Graphs

Graph condensation has emerged as an intriguing technique to provide Graph Neural Networks for large-scale graphs with a more compact yet informative small graph to save the expensive costs of large-scale graph learning. Despite the promising results achieved, previous graph condensation methods often employ an entangled condensation strategy that involves condensing nodes and edges simultaneously, leading to substantial GPU memory demands. This entangled strategy has considerably impeded the scalability of graph condensation, impairing its capability to condense extremely large-scale graphs and produce condensed graphs with high fidelity. Therefore, this paper presents Disentangled Condensation for large-scale graphs, abbreviated as DisCo, to provide scalable graph condensation for graphs of varying sizes. At the heart of DisCo are two complementary components, namely node and edge condensation modules, that realize the condensation of nodes and edges in a disentangled manner. In the node condensation module, we focus on synthesizing condensed nodes that exhibit a similar node feature distribution to original nodes using a pre-trained node classification model while incorporating class centroid alignment and anchor attachment regularizers. After node condensation, in the edge condensation module, we preserve the topology structure by transferring the link prediction model of the original graph to the condensed nodes, generating the corresponding condensed edges. Based on the disentangled strategy, the proposed DisCo can successfully scale up to the ogbn-papers100M graph with over 100 million nodes and 1 billion edges with flexible reduction rates. Extensive experiments on five common datasets further demonstrate that the proposed DisCo yields results superior to state-of-the-art counterparts by a significant margin. The source code is available at https://github.com/BangHonor/DisCo.