ZigzagAttention: Efficient Long-Context Inference with Exclusive Retrieval and Streaming Heads

With the rapid development of large language models (LLMs), handling long context has become one of the vital abilities in LLMs. Such long-context ability is accompanied by difficulties in deployment, especially due to the increased consumption of KV cache. There is certain work aiming to optimize the memory footprint of KV cache, inspired by the observation that attention heads can be categorized into retrieval heads that are of great significance and streaming heads that are of less significance. Typically, identifying the streaming heads and and waiving the KV cache in the streaming heads would largely reduce the overhead without hurting the performance that much. However, since employing both retrieval and streaming heads in one layer decomposes one large round of attention computation into two small ones, it may unexpectedly bring extra latency on accessing and indexing tensors. Based on this intuition, we impose an important improvement to the identification process of retrieval and streaming heads, in which we design a criterion that enforces exclusively retrieval or streaming heads gathered in one unique layer. In this way, we further eliminate the extra latency and only incur negligible performance degradation. Our method named \textsc{ZigzagAttention} is competitive among considered baselines owing to reduced latency and comparable performance.

MALM: A Multi-Information Adapter for Large Language Models to Mitigate Hallucination

Large language models (LLMs) are prone to three types of hallucination: Input-Conflicting, Context-Conflicting and Fact-Conflicting hallucinations. The purpose of this study is to mitigate the different types of hallucination by exploiting the interdependence between them. For this purpose, we propose a Multi-Information Adapter for Large Language Models (MALM). This framework employs a tailored multi-graph learning approach designed to elucidate the interconnections between original inputs, contextual information, and external factual knowledge, thereby alleviating the three categories of hallucination within a cohesive framework. Experiments were carried out on four benchmarking datasets: HaluEval, TruthfulQA, Natural Questions, and TriviaQA. We evaluated the proposed framework in two aspects: (1) adaptability to different base LLMs on HaluEval and TruthfulQA, to confirm if MALM is effective when applied on 7 typical LLMs. MALM showed significant improvements over LLaMA-2; (2) generalizability to retrieval-augmented generation (RAG) by combining MALM with three representative retrievers (BM25, Spider and DPR) separately. Furthermore, automated and human evaluations were conducted to substantiate the correctness of experimental results, where GPT-4 and 3 human volunteers judged which response was better between LLaMA-2 and MALM. The results showed that both GPT-4 and human preferred MALM in 79.4% and 65.6% of cases respectively. The results validate that incorporating the complex interactions between the three types of hallucination through a multilayered graph attention network into the LLM generation process is effective to mitigate the them. The adapter design of the proposed approach is also proven flexible and robust across different base LLMs.

MoDification: Mixture of Depths Made Easy

Long-context efficiency has recently become a trending topic in serving large language models (LLMs). And mixture of depths (MoD) is proposed as a perfect fit to bring down both latency and memory. In this paper, however, we discover that MoD can barely transform existing LLMs without costly training over an extensive number of tokens. To enable the transformations from any LLMs to MoD ones, we showcase top-k operator in MoD should be promoted to threshold-p operator, and refinement to architecture and data should also be crafted along. All these designs form our method termed MoDification. Through a comprehensive set of experiments covering model scales from 3B to 70B, we exhibit MoDification strikes an excellent balance between efficiency and effectiveness. MoDification can achieve up to ~1.2x speedup in latency and ~1.8x reduction in memory compared to original LLMs especially in long-context applications.

Investigating Context Effects in Similarity Judgements in Large Language Models

Large Language Models (LLMs) have revolutionised the capability of AI models in comprehending and generating natural language text. They are increasingly being used to empower and deploy agents in real-world scenarios, which make decisions and take actions based on their understanding of the context. Therefore researchers, policy makers and enterprises alike are working towards ensuring that the decisions made by these agents align with human values and user expectations. That being said, human values and decisions are not always straightforward to measure and are subject to different cognitive biases. There is a vast section of literature in Behavioural Science which studies biases in human judgements. In this work we report an ongoing investigation on alignment of LLMs with human judgements affected by order bias. Specifically, we focus on a famous human study which showed evidence of order effects in similarity judgements, and replicate it with various popular LLMs. We report the different settings where LLMs exhibit human-like order effect bias and discuss the implications of these findings to inform the design and development of LLM based applications.

Bi-DCSpell: A Bi-directional Detector-Corrector Interactive Framework for Chinese Spelling Check

Chinese Spelling Check (CSC) aims to detect and correct potentially misspelled characters in Chinese sentences. Naturally, it involves the detection and correction subtasks, which interact with each other dynamically. Such interactions are bi-directional, i.e., the detection result would help reduce the risk of over-correction and under-correction while the knowledge learnt from correction would help prevent false detection. Current CSC approaches are of two types: correction-only or single-directional detection-to-correction interactive frameworks. Nonetheless, they overlook the bi-directional interactions between detection and correction. This paper aims to fill the gap by proposing a Bi-directional Detector-Corrector framework for CSC (Bi-DCSpell). Notably, Bi-DCSpell contains separate detection and correction encoders, followed by a novel interactive learning module facilitating bi-directional feature interactions between detection and correction to improve each other's representation learning. Extensive experimental results demonstrate a robust correction performance of Bi-DCSpell on widely used benchmarking datasets while possessing a satisfactory detection ability.

A Supervised Information Enhanced Multi-Granularity Contrastive Learning Framework for EEG Based Emotion Recognition

This study introduces a novel Supervised Info-enhanced Contrastive Learning framework for EEG based Emotion Recognition (SICLEER). SI-CLEER employs multi-granularity contrastive learning to create robust EEG contextual representations, potentiallyn improving emotion recognition effectiveness. Unlike existing methods solely guided by classification loss, we propose a joint learning model combining self-supervised contrastive learning loss and supervised classification loss. This model optimizes both loss functions, capturing subtle EEG signal differences specific to emotion detection. Extensive experiments demonstrate SI-CLEER's robustness and superior accuracy on the SEED dataset compared to state-of-the-art methods. Furthermore, we analyze electrode performance, highlighting the significance of central frontal and temporal brain region EEGs in emotion detection. This study offers an universally applicable approach with potential benefits for diverse EEG classification tasks.

Beyond the Speculative Game: A Survey of Speculative Execution in Large Language Models

With the increasingly giant scales of (causal) large language models (LLMs), the inference efficiency comes as one of the core concerns along the improved performance. In contrast to the memory footprint, the latency bottleneck seems to be of greater importance as there can be billions of requests to a LLM (e.g., GPT-4) per day. The bottleneck is mainly due to the autoregressive innateness of LLMs, where tokens can only be generated sequentially during decoding. To alleviate the bottleneck, the idea of speculative execution, which originates from the field of computer architecture, is introduced to LLM decoding in a \textit{draft-then-verify} style. Under this regime, a sequence of tokens will be drafted in a fast pace by utilizing some heuristics, and then the tokens shall be verified in parallel by the LLM. As the costly sequential inference is parallelized, LLM decoding speed can be significantly boosted. Driven by the success of LLMs in recent couple of years, a growing literature in this direction has emerged. Yet, there lacks a position survey to summarize the current landscape and draw a roadmap for future development of this promising area. To meet this demand, we present the very first survey paper that reviews and unifies literature of speculative execution in LLMs (e.g., blockwise parallel decoding, speculative decoding, etc.) in a comprehensive framework and a systematic taxonomy. Based on the taxonomy, we present a critical review and comparative analysis of the current arts. Finally we highlight various key challenges and future directions to further develop the area.

LLM-Oriented Retrieval Tuner

Dense Retrieval (DR) is now considered as a promising tool to enhance the memorization capacity of Large Language Models (LLM) such as GPT3 and GPT-4 by incorporating external memories. However, due to the paradigm discrepancy between text generation of LLM and DR, it is still an open challenge to integrate the retrieval and generation tasks in a shared LLM. In this paper, we propose an efficient LLM-Oriented Retrieval Tuner, namely LMORT, which decouples DR capacity from base LLM and non-invasively coordinates the optimally aligned and uniform layers of the LLM towards a unified DR space, achieving an efficient and effective DR without tuning the LLM itself. The extensive experiments on six BEIR datasets show that our approach could achieve competitive zero-shot retrieval performance compared to a range of strong DR models while maintaining the generation ability of LLM.

Towards the Law of Capacity Gap in Distilling Language Models

Language model (LM) distillation is a trending area that aims to distil the knowledge resided in a large teacher LM to a small student one. While various methods have been proposed to push the distillation to its limits, it is still a pain distilling LMs when a large capacity gap is exhibited between the teacher and the student LMs. The pain is mainly resulted by the curse of capacity gap, which describes that a larger teacher LM cannot always lead to a better student LM than one distilled from a smaller teacher LM due to the affect of capacity gap increment. That is, there is likely an optimal point yielding the best student LM along the scaling course of the teacher LM. Even worse, the curse of capacity gap can be only partly yet not fully lifted as indicated in previous studies. However, the tale is not ever one-sided. Although a larger teacher LM has better performance than a smaller teacher LM, it is much more resource-demanding especially in the context of recent large LMs (LLMs). Consequently, instead of sticking to lifting the curse, leaving the curse as is should be arguably fine. Even better, in this paper, we reveal that the optimal capacity gap is almost consistent across different student scales and architectures, fortunately turning the curse into the law of capacity gap. The law later guides us to distil a 3B student LM (termed MiniMA) from a 7B teacher LM (adapted LLaMA2-7B). MiniMA is demonstrated to yield a new compute-performance pareto frontier among existing 3B LMs on commonly used benchmarks, and its instruction-tuned version (termed MiniChat) outperforms a wide range of 3B competitors in GPT4 evaluation and could even compete with several 7B chat models.

On Elastic Language Models

Large-scale pretrained language models have achieved compelling performance in a wide range of language understanding and information retrieval tasks. Knowledge distillation offers an opportunity to compress a large language model to a small one, in order to reach a reasonable latency-performance tradeoff. However, for scenarios where the number of requests (e.g., queries submitted to a search engine) is highly variant, the static tradeoff attained by the compressed language model might not always fit. Once a model is assigned with a static tradeoff, it could be inadequate in that the latency is too high when the number of requests is large or the performance is too low when the number of requests is small. To this end, we propose an elastic language model (ElasticLM) that elastically adjusts the tradeoff according to the request stream. The basic idea is to introduce a compute elasticity to the compressed language model, so that the tradeoff could vary on-the-fly along scalable and controllable compute. Specifically, we impose an elastic structure to enable ElasticLM with compute elasticity and design an elastic optimization to learn ElasticLM under compute elasticity. To serve ElasticLM, we apply an elastic schedule. Considering the specificity of information retrieval, we adapt ElasticLM to dense retrieval and reranking and present ElasticDenser and ElasticRanker respectively. Offline evaluation is conducted on a language understanding benchmark GLUE; and several information retrieval tasks including Natural Question, Trivia QA, and MS MARCO. The results show that ElasticLM along with ElasticDenser and ElasticRanker can perform correctly and competitively compared with an array of static baselines. Furthermore, online simulation with concurrency is also carried out. The results demonstrate that ElasticLM can provide elastic tradeoffs with respect to varying request stream.