CORD: Balancing COnsistency and Rank Distillation for Robust Retrieval-Augmented Generation

With the adoption of retrieval-augmented generation (RAG), large language models (LLMs) are expected to ground their generation to the retrieved contexts. Yet, this is hindered by position bias of LLMs, failing to evenly attend to all contexts. Previous work has addressed this by synthesizing contexts with perturbed positions of gold segment, creating a position-diversified train set. We extend this intuition to propose consistency regularization with augmentation and distillation. First, we augment each training instance with its position perturbation to encourage consistent predictions, regardless of ordering. We also distill behaviors of this pair, although it can be counterproductive in certain RAG scenarios where the given order from the retriever is crucial for generation quality. We thus propose CORD, balancing COnsistency and Rank Distillation. CORD adaptively samples noise-controlled perturbations from an interpolation space, ensuring both consistency and respect for the rank prior. Empirical results show this balance enables CORD to outperform consistently in diverse RAG benchmarks.

Inference Scaling for Bridging Retrieval and Augmented Generation

Retrieval-augmented generation (RAG) has emerged as a popular approach to steering the output of a large language model (LLM) by incorporating retrieved contexts as inputs. However, existing work observed the generator bias, such that improving the retrieval results may negatively affect the outcome. In this work, we show such bias can be mitigated, from inference scaling, aggregating inference calls from the permuted order of retrieved contexts. The proposed Mixture-of-Intervention (MOI) explicitly models the debiased utility of each passage with multiple forward passes to construct a new ranking. We also show that MOI can leverage the retriever's prior knowledge to reduce the computational cost by minimizing the number of permutations considered and lowering the cost per LLM call. We showcase the effectiveness of MOI on diverse RAG tasks, improving ROUGE-L on MS MARCO and EM on HotpotQA benchmarks by ~7 points.

SwiftKV: Fast Prefill-Optimized Inference with Knowledge-Preserving Model Transformation

LLM inference for popular enterprise use cases, such as summarization, RAG, and code-generation, typically observes orders of magnitude longer prompt lengths than generation lengths. This characteristic leads to high cost of prefill and increased response latency. In this paper, we present SwiftKV, a novel model transformation and distillation procedure specifically designed to reduce the time and cost of processing prompt tokens while preserving high quality of generated tokens. SwiftKV combines three key mechanisms: i) SingleInputKV, which prefills later layers' KV cache using a much earlier layer's output, allowing prompt tokens to skip much of the model computation, ii) AcrossKV, which merges the KV caches of neighboring layers to reduce the memory footprint and support larger batch size for higher throughput, and iii) a knowledge-preserving distillation procedure that can adapt existing LLMs for SwiftKV with minimal accuracy impact and low compute and data requirement. For Llama-3.1-8B and 70B, SwiftKV reduces the compute requirement of prefill by 50% and the memory requirement of the KV cache by 62.5% while incurring minimum quality degradation across a wide range of tasks. In the end-to-end inference serving using an optimized vLLM implementation, SwiftKV realizes up to 2x higher aggregate throughput and 60% lower time per output token. It can achieve a staggering 560 TFlops/GPU of normalized inference throughput, which translates to 16K tokens/s for Llama-3.1-70B in 16-bit precision on 4x H100 GPUs.

STUN: Structured-Then-Unstructured Pruning for Scalable MoE Pruning

Mixture-of-experts (MoEs) have been adopted for reducing inference costs by sparsely activating experts in Large language models (LLMs). Despite this reduction, the massive number of experts in MoEs still makes them expensive to serve. In this paper, we study how to address this, by pruning MoEs. Among pruning methodologies, unstructured pruning has been known to achieve the highest performance for a given pruning ratio, compared to structured pruning, since the latter imposes constraints on the sparsification structure. This is intuitive, as the solution space of unstructured pruning subsumes that of structured pruning. However, our counterintuitive finding reveals that expert pruning, a form of structured pruning, can actually precede unstructured pruning to outperform unstructured-only pruning. As existing expert pruning, requiring $O(\frac{k^n}{\sqrt{n}})$ forward passes for $n$ experts, cannot scale for recent MoEs, we propose a scalable alternative with $O(1)$ complexity, yet outperforming the more expensive methods. The key idea is leveraging a latent structure between experts, based on behavior similarity, such that the greedy decision of whether to prune closely captures the joint pruning effect. Ours is highly effective -- for Snowflake Arctic, a 480B-sized MoE with 128 experts, our method needs only one H100 and two hours to achieve nearly no loss in performance with 40% sparsity, even in generative tasks such as GSM8K, where state-of-the-art unstructured pruning fails to. The code will be made publicly available.

AI and Memory Wall

The availability of unprecedented unsupervised training data, along with neural scaling laws, has resulted in an unprecedented surge in model size and compute requirements for serving/training LLMs. However, the main performance bottleneck is increasingly shifting to memory bandwidth. Over the past 20 years, peak server hardware FLOPS has been scaling at 3.0x/2yrs, outpacing the growth of DRAM and interconnect bandwidth, which have only scaled at 1.6 and 1.4 times every 2 years, respectively. This disparity has made memory, rather than compute, the primary bottleneck in AI applications, particularly in serving. Here, we analyze encoder and decoder Transformer models and show how memory bandwidth can become the dominant bottleneck for decoder models. We argue for a redesign in model architecture, training, and deployment strategies to overcome this memory limitation.

Found in the Middle: How Language Models Use Long Contexts Better via Plug-and-Play Positional Encoding

This paper aims to overcome the "lost-in-the-middle" challenge of large language models (LLMs). While recent advancements have successfully enabled LLMs to perform stable language modeling with up to 4 million tokens, the persistent difficulty faced by most LLMs in identifying relevant information situated in the middle of the context has not been adequately tackled. To address this problem, this paper introduces Multi-scale Positional Encoding (Ms-PoE) which is a simple yet effective plug-and-play approach to enhance the capacity of LLMs to handle the relevant information located in the middle of the context, without fine-tuning or introducing any additional overhead. Ms-PoE leverages the position indice rescaling to relieve the long-term decay effect introduced by RoPE, while meticulously assigning distinct scaling ratios to different attention heads to preserve essential knowledge learned during the pre-training step, forming a multi-scale context fusion from short to long distance. Extensive experiments with a wide range of LLMs demonstrate the efficacy of our approach. Notably, Ms-PoE achieves an average accuracy gain of up to 3.8 on the Zero-SCROLLS benchmark over the original LLMs. Code are available at https://github.com/VITA-Group/Ms-PoE.

FP6-LLM: Efficiently Serving Large Language Models Through FP6-Centric Algorithm-System Co-Design

Six-bit quantization (FP6) can effectively reduce the size of large language models (LLMs) and preserve the model quality consistently across varied applications. However, existing systems do not provide Tensor Core support for FP6 quantization and struggle to achieve practical performance improvements during LLM inference. It is challenging to support FP6 quantization on GPUs due to (1) unfriendly memory access of model weights with irregular bit-width and (2) high runtime overhead of weight de-quantization. To address these problems, we propose TC-FPx, the first full-stack GPU kernel design scheme with unified Tensor Core support of float-point weights for various quantization bit-width. We integrate TC-FPx kernel into an existing inference system, providing new end-to-end support (called FP6-LLM) for quantized LLM inference, where better trade-offs between inference cost and model quality are achieved. Experiments show that FP6-LLM enables the inference of LLaMA-70b using only a single GPU, achieving 1.69x-2.65x higher normalized inference throughput than the FP16 baseline. The source code will be publicly available soon.

ZeroQuant(4+2): Redefining LLMs Quantization with a New FP6-Centric Strategy for Diverse Generative Tasks

This study examines 4-bit quantization methods like GPTQ in large language models (LLMs), highlighting GPTQ's overfitting and limited enhancement in Zero-Shot tasks. While prior works merely focusing on zero-shot measurement, we extend task scope to more generative categories such as code generation and abstractive summarization, in which we found that INT4 quantization can significantly underperform. However, simply shifting to higher precision formats like FP6 has been particularly challenging, thus overlooked, due to poor performance caused by the lack of sophisticated integration and system acceleration strategies on current AI hardware. Our results show that FP6, even with a coarse-grain quantization scheme, performs robustly across various algorithms and tasks, demonstrating its superiority in accuracy and versatility. Notably, with the FP6 quantization, \codestar-15B model performs comparably to its FP16 counterpart in code generation, and for smaller models like the 406M it closely matches their baselines in summarization. Neither can be achieved by INT4. To better accommodate various AI hardware and achieve the best system performance, we propose a novel 4+2 design for FP6 to achieve similar latency to the state-of-the-art INT4 fine-grain quantization. With our design, FP6 can become a promising solution to the current 4-bit quantization methods used in LLMs.

ZeroQuant-HERO: Hardware-Enhanced Robust Optimized Post-Training Quantization Framework for W8A8 Transformers

Quantization techniques are pivotal in reducing the memory and computational demands of deep neural network inference. Existing solutions, such as ZeroQuant, offer dynamic quantization for models like BERT and GPT but overlook crucial memory-bounded operators and the complexities of per-token quantization. Addressing these gaps, we present a novel, fully hardware-enhanced robust optimized post-training W8A8 quantization framework, ZeroQuant-HERO. This framework uniquely integrates both memory bandwidth and compute-intensive operators, aiming for optimal hardware performance. Additionally, it offers flexibility by allowing specific INT8 modules to switch to FP16/BF16 mode, enhancing accuracy.

DeepSpeed-VisualChat: Multi-Round Multi-Image Interleave Chat via Multi-Modal Causal Attention

Most of the existing multi-modal models, hindered by their incapacity to adeptly manage interleaved image-and-text inputs in multi-image, multi-round dialogues, face substantial constraints in resource allocation for training and data accessibility, impacting their adaptability and scalability across varied interaction realms. To address this, we present the DeepSpeed-VisualChat framework, designed to optimize Large Language Models (LLMs) by incorporating multi-modal capabilities, with a focus on enhancing the proficiency of Large Vision and Language Models in handling interleaved inputs. Our framework is notable for (1) its open-source support for multi-round and multi-image dialogues, (2) introducing an innovative multi-modal causal attention mechanism, and (3) utilizing data blending techniques on existing datasets to assure seamless interactions in multi-round, multi-image conversations. Compared to existing frameworks, DeepSpeed-VisualChat shows superior scalability up to 70B parameter language model size, representing a significant advancement in multi-modal language models and setting a solid foundation for future explorations.