High-quality prompts are crucial for Large Language Models (LLMs) to achieve exceptional performance. However, manually crafting effective prompts is labor-intensive and demands significant domain expertise, limiting its scalability. Existing automatic prompt optimization methods either extensively explore new prompt candidates, incurring high computational costs due to inefficient searches within a large solution space, or overly exploit feedback on existing prompts, risking suboptimal optimization because of the complex prompt landscape. To address these challenges, we introduce GreenTEA, an agentic LLM workflow for automatic prompt optimization that balances candidate exploration and knowledge exploitation. It leverages a collaborative team of agents to iteratively refine prompts based on feedback from error samples. An analyzing agent identifies common error patterns resulting from the current prompt via topic modeling, and a generation agent revises the prompt to directly address these key deficiencies. This refinement process is guided by a genetic algorithm framework, which simulates natural selection by evolving candidate prompts through operations such as crossover and mutation to progressively optimize model performance. Extensive numerical experiments conducted on public benchmark datasets suggest the superior performance of GreenTEA against human-engineered prompts and existing state-of-the-arts for automatic prompt optimization, covering logical and quantitative reasoning, commonsense, and ethical decision-making.

Delivering intelligent and adaptive navigation assistance in augmented reality (AR) requires more than visual cues, as it demands systems capable of interpreting flexible user intent and reasoning over both spatial and semantic context. Prior AR navigation systems often rely on rigid input schemes or predefined commands, which limit the utility of rich building data and hinder natural interaction. In this work, we propose an embodied AR navigation system that integrates Building Information Modeling (BIM) with a multi-agent retrieval-augmented generation (RAG) framework to support flexible, language-driven goal retrieval and route planning. The system orchestrates three language agents, Triage, Search, and Response, built on large language models (LLMs), which enables robust interpretation of open-ended queries and spatial reasoning using BIM data. Navigation guidance is delivered through an embodied AR agent, equipped with voice interaction and locomotion, to enhance user experience. A real-world user study yields a System Usability Scale (SUS) score of 80.5, indicating excellent usability, and comparative evaluations show that the embodied interface can significantly improves users' perception of system intelligence. These results underscore the importance and potential of language-grounded reasoning and embodiment in the design of user-centered AR navigation systems.

Multi-Head Latent Attention (MLA), introduced in DeepSeek-V2, compresses key-value states into a low-rank latent vector, caching only this vector to reduce memory. In tensor parallelism (TP), however, attention heads are computed across multiple devices, and each device must load the full cache, eroding the advantage of MLA over Grouped Query Attention (GQA). We propose Tensor-Parallel Latent Attention (TPLA): a scheme that partitions both the latent representation and each head's input dimension across devices, performs attention independently per shard, and then combines results with an all-reduce. TPLA preserves the benefits of a compressed KV cache while unlocking TP efficiency. Unlike Grouped Latent Attention (GLA), every head in TPLA still leverages the full latent representation, maintaining stronger representational capacity. TPLA is drop-in compatible with models pre-trained using MLA: it supports MLA-style prefilling and enables efficient tensor-parallel decoding without retraining. Applying simple orthogonal transforms -- e.g., the Hadamard transform or PCA -- before TP slicing further mitigates cross-shard interference, yielding minimal accuracy degradation. By reducing the per-device KV cache for DeepSeek-V3 and Kimi-K2, we achieve 1.79x and 1.93x speedups, respectively, at a 32K-token context length while maintaining performance on commonsense and LongBench benchmarks. TPLA can be implemented with FlashAttention-3, enabling practical end-to-end acceleration.

In recent years, a plethora of open-source foundation models have emerged, achieving remarkable progress in some widely attended fields, with performance being quite close to that of closed-source models. However, in high-value but more challenging scientific professional fields, either the fields still rely on expert models, or the progress of general foundation models lags significantly compared to those in popular areas, far from sufficient for transforming scientific research and leaving substantial gap between open-source models and closed-source models in these scientific domains. To mitigate this gap and explore a step further toward Artificial General Intelligence (AGI), we introduce Intern-S1, a specialized generalist equipped with general understanding and reasoning capabilities with expertise to analyze multiple science modal data. Intern-S1 is a multimodal Mixture-of-Experts (MoE) model with 28 billion activated parameters and 241 billion total parameters, continually pre-trained on 5T tokens, including over 2.5T tokens from scientific domains. In the post-training stage, Intern-S1 undergoes offline and then online reinforcement learning (RL) in InternBootCamp, where we propose Mixture-of-Rewards (MoR) to synergize the RL training on more than 1000 tasks simultaneously. Through integrated innovations in algorithms, data, and training systems, Intern-S1 achieved top-tier performance in online RL training. On comprehensive evaluation benchmarks, Intern-S1 demonstrates competitive performance on general reasoning tasks among open-source models and significantly outperforms open-source models in scientific domains, surpassing closed-source state-of-the-art models in professional tasks, such as molecular synthesis planning, reaction condition prediction, predicting thermodynamic stabilities for crystals. Our models are available at https://huggingface.co/internlm/Intern-S1.

Retrieval-Augmented Generation (RAG) has emerged as a standard framework for knowledge-intensive NLP tasks, combining large language models (LLMs) with document retrieval from external corpora. Despite its widespread use, most RAG pipelines continue to treat retrieval and reasoning as isolated components, retrieving documents once and then generating answers without further interaction. This static design often limits performance on complex tasks that require iterative evidence gathering or high-precision retrieval. Recent work in both the information retrieval (IR) and NLP communities has begun to close this gap by introducing adaptive retrieval and ranking methods that incorporate feedback. In this survey, we present a structured overview of advanced retrieval and ranking mechanisms that integrate such feedback. We categorize feedback signals based on their source and role in improving the query, retrieved context, or document pool. By consolidating these developments, we aim to bridge IR and NLP perspectives and highlight retrieval as a dynamic, learnable component of end-to-end RAG systems.

-- Speculative decoding has emerged as an effective approach for accelerating autoregressive inference by parallelizing token generation through a draft - then - verify paradigm. However, existing methods rely on static drafting lengths and rigid verification cri teria, limiting their adaptability across varying model uncertainties and input complexities. This paper proposes an information - theoretic framework for speculative decoding based on confidence - modulated drafting. By leveraging entropy and margin - based uncertainty measures over the drafter's output distribution, the proposed method dynamically adjusts the number of speculatively generated tokens at each iteration. This adaptive mechanism reduces rollback frequency, improves resource utilization, an d maintains output fidelity. Additionally, the verification process is modulated using the same confidence signals, enabling more flexible acceptance of drafted tokens without sacrificing generation quality. Experiments on machine translation and summariza tion tasks demonstrate significant speedups over standard speculative decoding while preserving or improving BLEU and ROUGE scores. The proposed approach offers a principled, plug - in method for efficient and robust decoding in large language models under v arying conditions of uncertainty. Keywords -- Speculative Decoding, Autoregressive Models, Confidence Estimation, Adaptive Inference, Entropy - Based Drafting, Sequence Generation, Large Language Models, Large Language Models (LLMs), Information - Theoretic Decoding. The task of sequence generation lies at the heart of numerous applications in natural language processing, including machine translation, text summarization, dialogue generation, and code synthesis. In the overwhelming majority of these applications, autor egressive (AR) decoding remains the dominant paradigm for generating sequences from a probabilistic language model [1 - 2] . Autoregressive models, particularly those based on the Transformer architecture, operate by predicting each token conditioned on the e ntire history of previously generated tokens. This left - to - right decoding strategy, though optimal in terms of likelihood estimation, suffers from a fundamental limitation: the inherently sequential nature of generation prohibits efficient parallelization, severely hindering inference throughput, especially in latency - sensitive deployment scenarios.

Large language models (LLMs) have demonstrated potential in educational applications, yet their capacity to accurately assess the cognitive alignment of reading materials with students' developmental stages remains insufficiently explored. This gap is particularly critical given the foundational educational principle of the Zone of Proximal Development (ZPD), which emphasizes the need to match learning resources with Students' Cognitive Abilities (SCA). Despite the importance of this alignment, there is a notable absence of comprehensive studies investigating LLMs' ability to evaluate reading comprehension difficulty across different student age groups, especially in the context of Chinese language education. To fill this gap, we introduce ZPD-SCA, a novel benchmark specifically designed to assess stage-level Chinese reading comprehension difficulty. The benchmark is annotated by 60 Special Grade teachers, a group that represents the top 0.15% of all in-service teachers nationwide. Experimental results reveal that LLMs perform poorly in zero-shot learning scenarios, with Qwen-max and GLM even falling below the probability of random guessing. When provided with in-context examples, LLMs performance improves substantially, with some models achieving nearly double the accuracy of their zero-shot baselines. These results reveal that LLMs possess emerging abilities to assess reading difficulty, while also exposing limitations in their current training for educationally aligned judgment. Notably, even the best-performing models display systematic directional biases, suggesting difficulties in accurately aligning material difficulty with SCA. Furthermore, significant variations in model performance across different genres underscore the complexity of task. We envision that ZPD-SCA can provide a foundation for evaluating and improving LLMs in cognitively aligned educational applications.

Tool-augmented large language models (LLMs) leverage external functions to extend their capabilities, but inaccurate function calls can lead to inefficiencies and increased costs.Existing methods address this challenge by fine-tuning LLMs or using demonstration-based prompting, yet they often suffer from high training overhead and fail to account for inconsistent demonstration samples, which misguide the model's invocation behavior. In this paper, we trained a behavior-aligned retriever (BAR), which provides behaviorally consistent demonstrations to help LLMs make more accurate tool-using decisions. To train the BAR, we construct a corpus including different function-calling behaviors, i.e., calling or non-calling.We use the contrastive learning framework to train the BAR with customized positive/negative pairs and a dual-negative contrastive loss, ensuring robust retrieval of behaviorally consistent examples.Experiments demonstrate that our approach significantly reduces erroneous function calls while maintaining high task performance, offering a cost-effective and efficient solution for tool-augmented LLMs.

As researchers increasingly adopt LLMs as writing assistants, generating high-quality research paper introductions remains both challenging and essential. We introduce Scientific Introduction Generation (SciIG), a task that evaluates LLMs' ability to produce coherent introductions from titles, abstracts, and related works. Curating new datasets from NAACL 2025 and ICLR 2025 papers, we assess five state-of-the-art models, including both open-source (DeepSeek-v3, Gemma-3-12B, LLaMA 4-Maverick, MistralAI Small 3.1) and closed-source GPT-4o systems, across multiple dimensions: lexical overlap, semantic similarity, content coverage, faithfulness, consistency, citation correctness, and narrative quality. Our comprehensive framework combines automated metrics with LLM-as-a-judge evaluations. Results demonstrate LLaMA-4 Maverick's superior performance on most metrics, particularly in semantic similarity and faithfulness. Moreover, three-shot prompting consistently outperforms fewer-shot approaches. These findings provide practical insights into developing effective research writing assistants and set realistic expectations for LLM-assisted academic writing. To foster reproducibility and future research, we will publicly release all code and datasets.

Diffusion-based Large Language Models (dLLMs) parallelize text generation by framing decoding as a denoising process, but suffer from high computational overhead since they predict all future suffix tokens at each step while retaining only a small fraction. We propose Diffusion Scratchpad (DPad), a training-free method that restricts attention to a small set of nearby suffix tokens, preserving fidelity while eliminating redundancy. DPad integrates two strategies: (i) a sliding window, which maintains a fixed-length suffix window, and (ii) distance-decay dropout, which deterministically removes distant suffix tokens before attention computation. This simple design is compatible with existing optimizations such as prefix caching and can be implemented with only a few lines of code. Comprehensive evaluations across multiple benchmarks on LLaDA-1.5 and Dream models demonstrate that DPad delivers up to $\mathbf{61.4\times}$ speedup over vanilla dLLMs while maintaining comparable accuracy, highlighting its potential for efficient and scalable long-sequence inference. Our code is available at https://github.com/Crys-Chen/DPad.