Xingcheng Zhang

Pan Zhang, Xiaoyi Dong, Bin Wang et al. · pku

We propose InternLM-XComposer, a vision-language large model that enables advanced image-text comprehension and composition. The innovative nature of our model is highlighted by three appealing properties: 1) Interleaved Text-Image Composition: InternLM-XComposer can effortlessly generate coherent and contextual articles that seamlessly integrate images, providing a more engaging and immersive reading experience. Simply provide a writing instruction, and our system will generate the corresponding manuscript. It can intelligently identify the areas in the text where images would enhance the content and automatically insert the most appropriate visual candidates. 2) Comprehension with Rich Multilingual Knowledge: The text-image comprehension is empowered by training on an extensive multi-modal multilingual database with carefully crafted strategies, resulting in a deep understanding of visual content. 3) State-of-the-art Performance: Our model consistently achieves state-of-the-art results across various mainstream benchmarks for vision-language foundational models, including MME Benchmark, MMBench, MMBench-CN, Seed-Bench, CCBench (Chinese Cultural Benchmark), QBench and Tiny LVLM. Owing to the absence of established metrics for quantitatively assessing text-image composition, we have devised a robust evaluation procedure that comprises both human and GPT4-Vision (GPT4-V) to ensure reliability. Notably, our InternLM-XComposer achieves competitive text-image composition scores compared to public solutions, including GPT4-V and GPT3.5. Collectively, InternLM-XComposer seamlessly blends advanced text-image comprehension and composition, revolutionizing vision-language interaction and offering new insights and opportunities. The InternLM-XComposer model series are publicly available at https://github.com/InternLM/InternLM-XComposer.

Pan Zhang, Xiaoyi Dong, Yuhang Zang et al. · pku

We present InternLM-XComposer-2.5 (IXC-2.5), a versatile large-vision language model that supports long-contextual input and output. IXC-2.5 excels in various text-image comprehension and composition applications, achieving GPT-4V level capabilities with merely 7B LLM backend. Trained with 24K interleaved image-text contexts, it can seamlessly extend to 96K long contexts via RoPE extrapolation. This long-context capability allows IXC-2.5 to excel in tasks requiring extensive input and output contexts. Compared to its previous 2.0 version, InternLM-XComposer-2.5 features three major upgrades in vision-language comprehension: (1) Ultra-High Resolution Understanding, (2) Fine-Grained Video Understanding, and (3) Multi-Turn Multi-Image Dialogue. In addition to comprehension, IXC-2.5 extends to two compelling applications using extra LoRA parameters for text-image composition: (1) Crafting Webpages and (2) Composing High-Quality Text-Image Articles. IXC-2.5 has been evaluated on 28 benchmarks, outperforming existing open-source state-of-the-art models on 16 benchmarks. It also surpasses or competes closely with GPT-4V and Gemini Pro on 16 key tasks. The InternLM-XComposer-2.5 is publicly available at https://github.com/InternLM/InternLM-XComposer.

Fan Cui, Chenyang Yin, Kexing Zhou et al.

Recent studies have demonstrated the significant potential of Large Language Models (LLMs) in generating Register Transfer Level (RTL) code, with notable advancements showcased by commercial models such as GPT-4 and Claude3-Opus. However, these proprietary LLMs often raise concerns regarding privacy and security. While open-source LLMs offer solutions to these concerns, they typically underperform commercial models in RTL code generation tasks, primarily due to the scarcity of high-quality open-source RTL datasets. To address this challenge, we introduce OriGen , a fully open-source framework that incorporates self-reflection capabilities and a novel dataset augmentation methodology for generating high-quality, large-scale RTL code. Our approach employs a code-tocode augmentation technique to enhance the quality of open-source RTL code datasets. Furthermore, OriGen can rectify syntactic errors through a self-reflection process that leverages compiler feedback. Experimental results demonstrate that OriGen significantly outperforms other open-source alternatives in RTL code generation. It surpasses the previous best-performing open-source LLM by 12.8% and even exceeds GPT-4 Turbo in the pass@1 metric on the VerilogEval-Human benchmark. Moreover, OriGen exhibits superior capabilities in self-reflection and error correction, outperforming GPT-4 by 19.9% on a benchmark designed to evaluate self-reflection capabilities.

Guofeng Feng, Siyan Chen, Rong Fu et al.

This work introduces FlashGS, an open-source CUDA Python library, designed to facilitate the efficient differentiable rasterization of 3D Gaussian Splatting through algorithmic and kernel-level optimizations. FlashGS is developed based on the observations from a comprehensive analysis of the rendering process to enhance computational efficiency and bring the technique to wide adoption. The paper includes a suite of optimization strategies, encompassing redundancy elimination, efficient pipelining, refined control and scheduling mechanisms, and memory access optimizations, all of which are meticulously integrated to amplify the performance of the rasterization process. An extensive evaluation of FlashGS' performance has been conducted across a diverse spectrum of synthetic and real-world large-scale scenes, encompassing a variety of image resolutions. The empirical findings demonstrate that FlashGS consistently achieves an average 4x acceleration over mobile consumer GPUs, coupled with reduced memory consumption. These results underscore the superior performance and resource optimization capabilities of FlashGS, positioning it as a formidable tool in the domain of 3D rendering.

Zhe Chen, Weiyun Wang, Yue Cao et al.

We introduce InternVL 2.5, an advanced multimodal large language model (MLLM) series that builds upon InternVL 2.0, maintaining its core model architecture while introducing significant enhancements in training and testing strategies as well as data quality. In this work, we delve into the relationship between model scaling and performance, systematically exploring the performance trends in vision encoders, language models, dataset sizes, and test-time configurations. Through extensive evaluations on a wide range of benchmarks, including multi-discipline reasoning, document understanding, multi-image / video understanding, real-world comprehension, multimodal hallucination detection, visual grounding, multilingual capabilities, and pure language processing, InternVL 2.5 exhibits competitive performance, rivaling leading commercial models such as GPT-4o and Claude-3.5-Sonnet. Notably, our model is the first open-source MLLMs to surpass 70% on the MMMU benchmark, achieving a 3.7-point improvement through Chain-of-Thought (CoT) reasoning and showcasing strong potential for test-time scaling. We hope this model contributes to the open-source community by setting new standards for developing and applying multimodal AI systems. HuggingFace demo see https://huggingface.co/spaces/OpenGVLab/InternVL

Jinguo Zhu, Weiyun Wang, Zhe Chen et al.

We introduce InternVL3, a significant advancement in the InternVL series featuring a native multimodal pre-training paradigm. Rather than adapting a text-only large language model (LLM) into a multimodal large language model (MLLM) that supports visual inputs, InternVL3 jointly acquires multimodal and linguistic capabilities from both diverse multimodal data and pure-text corpora during a single pre-training stage. This unified training paradigm effectively addresses the complexities and alignment challenges commonly encountered in conventional post-hoc training pipelines for MLLMs. To further improve performance and scalability, InternVL3 incorporates variable visual position encoding (V2PE) to support extended multimodal contexts, employs advanced post-training techniques such as supervised fine-tuning (SFT) and mixed preference optimization (MPO), and adopts test-time scaling strategies alongside an optimized training infrastructure. Extensive empirical evaluations demonstrate that InternVL3 delivers superior performance across a wide range of multi-modal tasks. In particular, InternVL3-78B achieves a score of 72.2 on the MMMU benchmark, setting a new state-of-the-art among open-source MLLMs. Its capabilities remain highly competitive with leading proprietary models, including ChatGPT-4o, Claude 3.5 Sonnet, and Gemini 2.5 Pro, while also maintaining strong pure-language proficiency. In pursuit of open-science principles, we will publicly release both the training data and model weights to foster further research and development in next-generation MLLMs.

Xiaoyi Dong, Pan Zhang, Yuhang Zang et al. · pku

We introduce InternLM-XComposer2, a cutting-edge vision-language model excelling in free-form text-image composition and comprehension. This model goes beyond conventional vision-language understanding, adeptly crafting interleaved text-image content from diverse inputs like outlines, detailed textual specifications, and reference images, enabling highly customizable content creation. InternLM-XComposer2 proposes a Partial LoRA (PLoRA) approach that applies additional LoRA parameters exclusively to image tokens to preserve the integrity of pre-trained language knowledge, striking a balance between precise vision understanding and text composition with literary talent. Experimental results demonstrate the superiority of InternLM-XComposer2 based on InternLM2-7B in producing high-quality long-text multi-modal content and its exceptional vision-language understanding performance across various benchmarks, where it not only significantly outperforms existing multimodal models but also matches or even surpasses GPT-4V and Gemini Pro in certain assessments. This highlights its remarkable proficiency in the realm of multimodal understanding. The InternLM-XComposer2 model series with 7B parameters are publicly available at https://github.com/InternLM/InternLM-XComposer.

Zheng Cai, Maosong Cao, Haojiong Chen et al. · pku

The evolution of Large Language Models (LLMs) like ChatGPT and GPT-4 has sparked discussions on the advent of Artificial General Intelligence (AGI). However, replicating such advancements in open-source models has been challenging. This paper introduces InternLM2, an open-source LLM that outperforms its predecessors in comprehensive evaluations across 6 dimensions and 30 benchmarks, long-context modeling, and open-ended subjective evaluations through innovative pre-training and optimization techniques. The pre-training process of InternLM2 is meticulously detailed, highlighting the preparation of diverse data types including text, code, and long-context data. InternLM2 efficiently captures long-term dependencies, initially trained on 4k tokens before advancing to 32k tokens in pre-training and fine-tuning stages, exhibiting remarkable performance on the 200k ``Needle-in-a-Haystack" test. InternLM2 is further aligned using Supervised Fine-Tuning (SFT) and a novel Conditional Online Reinforcement Learning from Human Feedback (COOL RLHF) strategy that addresses conflicting human preferences and reward hacking. By releasing InternLM2 models in different training stages and model sizes, we provide the community with insights into the model's evolution.

Zimu Liao, Siyan Chen, Rong Fu et al.

Recently, 3D Gaussian Splatting (3DGS) has garnered attention for its high fidelity and real-time rendering. However, adapting 3DGS to different camera models, particularly fisheye lenses, poses challenges due to the unique 3D to 2D projection calculation. Additionally, there are inefficiencies in the tile-based splatting, especially for the extreme curvature and wide field of view of fisheye lenses, which are crucial for its broader real-life applications. To tackle these challenges, we introduce Fisheye-GS.This innovative method recalculates the projection transformation and its gradients for fisheye cameras. Our approach can be seamlessly integrated as a module into other efficient 3D rendering methods, emphasizing its extensibility, lightweight nature, and modular design. Since we only modified the projection component, it can also be easily adapted for use with different camera models. Compared to methods that train after undistortion, our approach demonstrates a clear improvement in visual quality.

Xiaoyi Dong, Pan Zhang, Yuhang Zang et al. · pku

The Large Vision-Language Model (LVLM) field has seen significant advancements, yet its progression has been hindered by challenges in comprehending fine-grained visual content due to limited resolution. Recent efforts have aimed to enhance the high-resolution understanding capabilities of LVLMs, yet they remain capped at approximately 1500 x 1500 pixels and constrained to a relatively narrow resolution range. This paper represents InternLM-XComposer2-4KHD, a groundbreaking exploration into elevating LVLM resolution capabilities up to 4K HD (3840 x 1600) and beyond. Concurrently, considering the ultra-high resolution may not be necessary in all scenarios, it supports a wide range of diverse resolutions from 336 pixels to 4K standard, significantly broadening its scope of applicability. Specifically, this research advances the patch division paradigm by introducing a novel extension: dynamic resolution with automatic patch configuration. It maintains the training image aspect ratios while automatically varying patch counts and configuring layouts based on a pre-trained Vision Transformer (ViT) (336 x 336), leading to dynamic training resolution from 336 pixels to 4K standard. Our research demonstrates that scaling training resolution up to 4K HD leads to consistent performance enhancements without hitting the ceiling of potential improvements. InternLM-XComposer2-4KHD shows superb capability that matches or even surpasses GPT-4V and Gemini Pro in 10 of the 16 benchmarks. The InternLM-XComposer2-4KHD model series with 7B parameters are publicly available at https://github.com/InternLM/InternLM-XComposer.

Shiguang Wang, Tao Xie, Haijun Liu et al.

Channel Pruning is one of the most widespread techniques used to compress deep neural networks while maintaining their performances. Currently, a typical pruning algorithm leverages neural architecture search to directly find networks with a configurable width, the key step of which is to identify representative subnet for various pruning ratios by training a supernet. However, current methods mainly follow a serial training strategy to optimize supernet, which is very time-consuming. In this work, we introduce PSE-Net, a novel parallel-subnets estimator for efficient channel pruning. Specifically, we propose a parallel-subnets training algorithm that simulate the forward-backward pass of multiple subnets by droping extraneous features on batch dimension, thus various subnets could be trained in one round. Our proposed algorithm facilitates the efficiency of supernet training and equips the network with the ability to interpolate the accuracy of unsampled subnets, enabling PSE-Net to effectively evaluate and rank the subnets. Over the trained supernet, we develop a prior-distributed-based sampling algorithm to boost the performance of classical evolutionary search. Such algorithm utilizes the prior information of supernet training phase to assist in the search of optimal subnets while tackling the challenge of discovering samples that satisfy resource constraints due to the long-tail distribution of network configuration. Extensive experiments demonstrate PSE-Net outperforms previous state-of-the-art channel pruning methods on the ImageNet dataset while retaining superior supernet training efficiency. For example, under 300M FLOPs constraint, our pruned MobileNetV2 achieves 75.2% Top-1 accuracy on ImageNet dataset, exceeding the original MobileNetV2 by 2.6 units while only cost 30%/16% times than BCNet/AutoAlim.

Jiefei Chen, Binbin Lin, Jinming Ma et al.

Modern serving systems for Mixture-of-Experts (MoE) models adopt hybrid data-expert parallelism: expert parallelism (EP) shards experts across GPUs to scale capacity, while data parallelism (DP) replicates attention layers across instances to process independent requests. Existing systems bind each request's attention, MoE communication, and KV cache to a single instance. Because attention latency scales with KV cache size while MoE communication latency scales with batch size, this binding cannot balance both simultaneously, producing EP stragglers; it also fragments KV memory across instances, inflating tail latency under long contexts. While existing context parallelism (CP) mitigates these constraints, its uniform parallelism degree incurs prohibitive communication and attention-side overheads. We present \work, which decouples MoE communication from KV cache placement and achieves dual balance through dynamic context parallelism (DCP). DCP assigns each request a context-parallel degree sized to its KV footprint: long requests distribute attention across multiple instances; short requests remain local. This dynamic parallelism effectively liquefies the KV cache across the cluster, balancing both the per-instance KV cache occupancy and batch sizes without unnecessary load-balancing costs. To bridge DCP with static execution, \work introduces an ahead-of-time (AOT) graph engine paired with a custom routing-based communication backend. Experimental results show that \work maintains up to $1.88\times$--$3.27\times$ higher request rates under strict time-per-output-token (TPOT) service level objectives (SLOs). Furthermore, \work significantly mitigates stragglers, reducing P99 tail latency by up to $1.79\times$--$2.12\times$.

Haojie Duanmu, Xiuhong Li, Zhihang Yuan et al.

Mixture-of-Experts (MoE) models face deployment challenges due to their large parameter counts and computational demands. We explore quantization for MoE models and highlight two key insights: 1) linear blocks exhibit varying quantization sensitivity, and 2) divergent expert activation frequencies create heterogeneous computational characteristics. Based on these observations, we introduce MxMoE, a mixed-precision optimization framework for MoE models that considers both algorithmic and system perspectives. MxMoE navigates the design space defined by parameter sensitivity, expert activation dynamics, and hardware resources to derive efficient mixed-precision configurations. Additionally, MxMoE automatically generates optimized mixed-precision GroupGEMM kernels, enabling parallel execution of GEMMs with different precisions. Evaluations show that MxMoE outperforms existing methods, achieving 2.4 lower Wikitext-2 perplexity than GPTQ at 2.25-bit and delivering up to 3.4x speedup over full precision, as well as up to 29.4% speedup over uniform quantization at equivalent accuracy with 5-bit weight-activation quantization. Our code is available at https://github.com/cat538/MxMoE.

Haoran Xu, Ziqian Liu, Rong Fu et al.

With the evolution of large language models, traditional Transformer models become computationally demanding for lengthy sequences due to the quadratic growth in computation with respect to the sequence length. Mamba, emerging as a groundbreaking architecture in the field of generative AI, demonstrates remarkable proficiency in handling elongated sequences with reduced computational and memory complexity. Nevertheless, the existing training framework of Mamba presents inefficiency with variable-length sequence inputs. Either single-sequence training results in low GPU utilization, or batched processing of variable-length sequences to a maximum length incurs considerable memory and computational overhead. To address this problem, we analyze the performance of bottleneck operators in Mamba under diverse tensor shapes and proposed PackMamba, a high-throughput Mamba that efficiently handles variable-length sequences. Diving deep into state-space models (SSMs), we modify the parallel operators to avoid passing information between individual sequences while maintaining high performance. Experimental results on an NVIDIA A100 GPU demonstrate throughput exceeding the baseline single-sequence processing scheme: 3.06x speedup on the 1.4B model and 2.62x on the 2.8B model.

Pan Zhang, Xiaoyi Dong, Yuhang Cao et al. · pku, tsinghua

Creating AI systems that can interact with environments over long periods, similar to human cognition, has been a longstanding research goal. Recent advancements in multimodal large language models (MLLMs) have made significant strides in open-world understanding. However, the challenge of continuous and simultaneous streaming perception, memory, and reasoning remains largely unexplored. Current MLLMs are constrained by their sequence-to-sequence architecture, which limits their ability to process inputs and generate responses simultaneously, akin to being unable to think while perceiving. Furthermore, relying on long contexts to store historical data is impractical for long-term interactions, as retaining all information becomes costly and inefficient. Therefore, rather than relying on a single foundation model to perform all functions, this project draws inspiration from the concept of the Specialized Generalist AI and introduces disentangled streaming perception, reasoning, and memory mechanisms, enabling real-time interaction with streaming video and audio input. The proposed framework InternLM-XComposer2.5-OmniLive (IXC2.5-OL) consists of three key modules: (1) Streaming Perception Module: Processes multimodal information in real-time, storing key details in memory and triggering reasoning in response to user queries. (2) Multi-modal Long Memory Module: Integrates short-term and long-term memory, compressing short-term memories into long-term ones for efficient retrieval and improved accuracy. (3) Reasoning Module: Responds to queries and executes reasoning tasks, coordinating with the perception and memory modules. This project simulates human-like cognition, enabling multimodal large language models to provide continuous and adaptive service over time.

Haojie Duanmu, Zhihang Yuan, Xiuhong Li et al.

Large language models (LLMs) can now handle longer sequences of tokens, enabling complex tasks like book understanding and generating lengthy novels. However, the key-value (KV) cache required for LLMs consumes substantial memory as context length increasing, becoming the bottleneck for deployment. In this paper, we present a strategy called SKVQ, which stands for sliding-window KV cache quantization, to address the issue of extremely low bitwidth KV cache quantization. To achieve this, SKVQ rearranges the channels of the KV cache in order to improve the similarity of channels in quantization groups, and applies clipped dynamic quantization at the group level. Additionally, SKVQ ensures that the most recent window tokens in the KV cache are preserved with high precision. This helps maintain the accuracy of a small but important portion of the KV cache.SKVQ achieves high compression ratios while maintaining accuracy. Our evaluation on LLMs demonstrates that SKVQ surpasses previous quantization approaches, allowing for quantization of the KV cache to 2-bit keys and 1.5-bit values with minimal loss of accuracy. With SKVQ, it is possible to process context lengths of up to 1M on an 80GB memory GPU for a 7b model and up to 7 times faster decoding.

Miao Tao, Yuanzhen Zhou, Haoran Xu et al.

Rendering large-scale 3D Gaussian Splatting (3DGS) model faces significant challenges in achieving real-time, high-fidelity performance on consumer-grade devices. Fully realizing the potential of 3DGS in applications such as virtual reality (VR) requires addressing critical system-level challenges to support real-time, immersive experiences. We propose GS-Cache, an end-to-end framework that seamlessly integrates 3DGS's advanced representation with a highly optimized rendering system. GS-Cache introduces a cache-centric pipeline to eliminate redundant computations, an efficiency-aware scheduler for elastic multi-GPU rendering, and optimized CUDA kernels to overcome computational bottlenecks. This synergy between 3DGS and system design enables GS-Cache to achieve up to 5.35x performance improvement, 35% latency reduction, and 42% lower GPU memory usage, supporting 2K binocular rendering at over 120 FPS with high visual quality. By bridging the gap between 3DGS's representation power and the demands of VR systems, GS-Cache establishes a scalable and efficient framework for real-time neural rendering in immersive environments.

Jiecheng Zhou, Ding Tang, Rong Fu et al.

The burgeoning computational demands for training large language models (LLMs) necessitate efficient methods, including quantized training, which leverages low-bit arithmetic operations to reduce costs. While FP8 precision has shown potential, leveraging FP4 remains challenging due to inherent quantization errors and limited representation capability. Based on the Transformer architecture, we present an FP4 training scheme for LLMs, overcoming these obstacles through mixed-precision quantization strategies tailed for different modules and training stages. This allows us to apply the precision level suitable to distinct components within the model, ensuring that multi-head attention and linear layers are handled appropriately. Our pretraining recipe ensures stability in backpropagation by incorporating fine-grained quantization methods with a target precision training schedule. Experimental results demonstrate that our FP4 training scheme achieves accuracy comparable to BF16 and FP8, with smaller theoretical computational cost. With the advent of next-generation hardware supporting FP4, our method sets the foundation for efficient ultra-low precision training.

Zimu Liao, Jifeng Ding, Siwei Cui et al.

3D Gaussian Splatting (3DGS) renders pixels by rasterizing Gaussian primitives, where conditional alpha-blending dominates the computational cost in the rendering pipeline. This paper proposes TC-GS, an algorithm-independent universal module that expands the applicability of Tensor Core (TCU) for 3DGS, leading to substantial speedups and seamless integration into existing 3DGS optimization frameworks. The key innovation lies in mapping alpha computation to matrix multiplication, fully utilizing otherwise idle TCUs in existing 3DGS implementations. TC-GS provides plug-and-play acceleration for existing top-tier acceleration algorithms and integrates seamlessly with rendering pipeline designs, such as Gaussian compression and redundancy elimination algorithms. Additionally, we introduce a global-to-local coordinate transformation to mitigate rounding errors from quadratic terms of pixel coordinates caused by Tensor Core half-precision computation. Extensive experiments demonstrate that our method maintains rendering quality while providing an additional 2.18x speedup over existing Gaussian acceleration algorithms, thereby achieving a total acceleration of up to 5.6x.

Jiecheng Zhou, Qinghao Hu, Yuyang Jin et al.

Large Language Models (LLMs) are now widely used across many domains. With their rapid development, Reinforcement Learning with Verifiable Rewards (RLVR) has surged in recent months to enhance their reasoning and understanding abilities. However, its complex data flows and diverse tasks pose substantial challenges to RL training systems, and there is limited understanding of RLVR from a system perspective. To thoroughly understand the system challenges introduced by RLVR, we present a characterization study of RLVR tasks in our LLM deployment. Specifically, we investigate the distribution and variation trends of workloads across different RL tasks across training steps. We identify issues such as GPU idling caused by skewed sequence length distribution, inefficient parallel strategies in dynamically varying workloads, inefficient data management mechanisms, and load imbalance. We describe our observations and call for further investigation into the remaining open challenges. Furthermore, we propose PolyTrace benchmark suite to conduct evaluation with realistic workloads, and a practical use case validates that PolyTrace benchmark suite exhibits 94.7% accuracy.

Yu Tang, Chenyu Wang, Yufan Zhang et al.

The further development of deep neural networks is hampered by the limited GPU memory resource. Therefore, the optimization of GPU memory resources is highly demanded. Swapping and recomputation are commonly applied to make better use of GPU memory in deep learning. However, as an emerging domain, several challenges remain:1)The efficiency of recomputation is limited for both static and dynamic methods. 2)Swapping requires offloading parameters manually, which incurs a great time cost. 3) There is no such dynamic and fine-grained method that involves tensor swapping together with tensor recomputation nowadays. To remedy the above issues, we propose a novel scheduler manager named DELTA(Dynamic tEnsor offLoad and recompuTAtion). To the best of our knowledge, we are the first to make a reasonable dynamic runtime scheduler on the combination of tensor swapping and tensor recomputation without user oversight. In DELTA, we propose a filter algorithm to select the optimal tensors to be released out of GPU memory and present a director algorithm to select a proper action for each of these tensors. Furthermore, prefetching and overlapping are deliberately considered to overcome the time cost caused by swapping and recomputing tensors. Experimental results show that DELTA not only saves 40%-70% of GPU memory, surpassing the state-of-the-art method to a great extent but also gets comparable convergence results as the baseline with acceptable time delay. Also, DELTA gains 2.04$\times$ maximum batchsize when training ResNet-50 and 2.25$\times$ when training ResNet-101 compared with the baseline. Besides, comparisons between the swapping cost and recomputation cost in our experiments demonstrate the importance of making a reasonable dynamic scheduler on tensor swapping and tensor recomputation, which refutes the arguments in some related work that swapping should be the first and best choice.

Kai Chen, Jiaqi Wang, Shuo Yang et al.

High-performance object detection relies on expensive convolutional networks to compute features, often leading to significant challenges in applications, e.g. those that require detecting objects from video streams in real time. The key to this problem is to trade accuracy for efficiency in an effective way, i.e. reducing the computing cost while maintaining competitive performance. To seek a good balance, previous efforts usually focus on optimizing the model architectures. This paper explores an alternative approach, that is, to reallocate the computation over a scale-time space. The basic idea is to perform expensive detection sparsely and propagate the results across both scales and time with substantially cheaper networks, by exploiting the strong correlations among them. Specifically, we present a unified framework that integrates detection, temporal propagation, and across-scale refinement on a Scale-Time Lattice. On this framework, one can explore various strategies to balance performance and cost. Taking advantage of this flexibility, we further develop an adaptive scheme with the detector invoked on demand and thus obtain improved tradeoff. On ImageNet VID dataset, the proposed method can achieve a competitive mAP 79.6% at 20 fps, or 79.0% at 62 fps as a performance/speed tradeoff.

Xingcheng Zhang, Lei Yang, Junjie Yan et al.

Massive classification, a classification task defined over a vast number of classes (hundreds of thousands or even millions), has become an essential part of many real-world systems, such as face recognition. Existing methods, including the deep networks that achieved remarkable success in recent years, were mostly devised for problems with a moderate number of classes. They would meet with substantial difficulties, e.g. excessive memory demand and computational cost, when applied to massive problems. We present a new method to tackle this problem. This method can efficiently and accurately identify a small number of "active classes" for each mini-batch, based on a set of dynamic class hierarchies constructed on the fly. We also develop an adaptive allocation scheme thereon, which leads to a better tradeoff between performance and cost. On several large-scale benchmarks, our method significantly reduces the training cost and memory demand, while maintaining competitive performance.

Xingcheng Zhang, Zhizhong Li, Chen Change Loy et al.

A number of studies have shown that increasing the depth or width of convolutional networks is a rewarding approach to improve the performance of image recognition. In our study, however, we observed difficulties along both directions. On one hand, the pursuit for very deep networks is met with a diminishing return and increased training difficulty; on the other hand, widening a network would result in a quadratic growth in both computational cost and memory demand. These difficulties motivate us to explore structural diversity in designing deep networks, a new dimension beyond just depth and width. Specifically, we present a new family of modules, namely the PolyInception, which can be flexibly inserted in isolation or in a composition as replacements of different parts of a network. Choosing PolyInception modules with the guidance of architectural efficiency can improve the expressive power while preserving comparable computational cost. The Very Deep PolyNet, designed following this direction, demonstrates substantial improvements over the state-of-the-art on the ILSVRC 2012 benchmark. Compared to Inception-ResNet-v2, it reduces the top-5 validation error on single crops from 4.9% to 4.25%, and that on multi-crops from 3.7% to 3.45%.