Related papers: DeepSeek-Coder-V2: Breaking the Barrier of Closed-…

Large language models (LLMs) have shown continuously improving multilingual capabilities, and even small-scale open-source models have demonstrated rapid performance enhancement. In this paper, we systematically explore the abilities of…

Existing encoder-free vision-language models (VLMs) are rapidly narrowing the performance gap with their encoder-based counterparts, highlighting the promising potential for unified multimodal systems with structural simplicity and…

We present a single-GPU (H100, bf16) evaluation of GPT-OSS-20B (Mixture-of-Experts; 20.9B total, approx. 3.61B active) against dense baselines Qwen3-32B and Yi-34B across multiple dimensions. We measure true time-to-first-token (TTFT),…

While Mixture of Experts (MoE) models achieve remarkable efficiency by activating only subsets of parameters, they suffer from high memory access costs during inference. Memory-layer architectures offer an appealing alternative with very…

The ViDoRe Benchmark V1 was approaching saturation with top models exceeding 90% nDCG@5, limiting its ability to discern improvements. ViDoRe Benchmark V2 introduces realistic, challenging retrieval scenarios via blind contextual querying,…

Code reasoning is a fundamental capability for large language models (LLMs) in the code domain. It involves understanding and predicting a program's execution behavior, such as determining the output for a given input or whether a specific…

Predominant LLMs focus on high-resource languages while leaving low-resource languages, particularly those in Southeast Asia (SEA), underrepresented. In addition, those models are general-purpose and pay limited attention to the e-commerce…

Sparsely-gated Mixture of Experts networks (MoEs) have demonstrated excellent scalability in Natural Language Processing. In Computer Vision, however, almost all performant networks are "dense", that is, every input is processed by every…

Code completion models have made significant progress in recent years. Recently, repository-level code completion has drawn more attention in modern software development, and several baseline methods and benchmarks have been proposed.…

Despite LLMs' excellent code creation capabilities, multilingual code generation remains extremely challenging. To address this, we intent to improve the multi-programming-lingual (MultiPL) performance of the base LLMs while retaining the…

The Mixture-of-Experts (MoE) model has emerged as a prominent architecture in the field of Large Language Models (LLMs), providing a better balance between model performance and computational efficiency. However the General Matrix Multiply…

Deep search has become a crucial capability for frontier multimodal agents, enabling models to solve complex questions through active search, evidence verification, and multi-step reasoning. Despite rapid progress, top-tier multimodal…

With the rapid advancement of video understanding, existing benchmarks are becoming increasingly saturated, exposing a critical discrepancy between inflated leaderboard scores and real-world model capabilities. To address this widening gap,…

Despite the success of text retrieval in many NLP tasks, code retrieval remains a largely underexplored area. Most text retrieval systems are tailored for natural language queries, often neglecting the specific challenges of retrieving…

Current movie dubbing technology can generate the desired voice from a given speech prompt, ensuring good synchronization between speech and visuals while accurately conveying the intended emotions. However, in movie dubbing, key aspects…

Efficient deployment of large language models, particularly Mixture of Experts (MoE), on resource-constrained platforms presents significant challenges, especially in terms of computational efficiency and memory utilization. The MoE…

Large Language and Vision Models using a Mixture-of-Experts (MoE) architecture pose significant challenges for deployment due to their computational and memory demands. Mixed Precision Quantization assigns different precisions to different…

Most vulnerability detection studies focus on datasets of vulnerabilities in C/C++ code, offering limited language diversity. Thus, the effectiveness of deep learning methods, including large language models (LLMs), in detecting software…

Mixture-of-Experts (MoE) is a neural network architecture that adds sparsely activated expert blocks to a base model, increasing the number of parameters without impacting computational costs. However, current distributed deep learning…

Code completion is one of the most useful features in the Integrated Development Environments (IDEs), which can accelerate software development by suggesting the next probable token based on the contextual code in real-time. Recent studies…