Large Language Models (LLMs) with vast context windows offer new avenues for in-context learning (ICL), where providing many examples ("many-shot" prompting) is often assumed to enhance performance. We investigate this assumption for the complex task of code translation. Through a large-scale empirical study of over 90,000 translations, we systematically evaluate the impact of scaling in-context examples from zero-shot to many-shot configurations of up to 625 examples, with prompts spanning from approximately 100,000 to 800,000 tokens. Our findings reveal a "many-shot paradox": while static similarity metrics may modestly improve with more examples, functional correctness consistently peaks with few-shot prompting (5-25 examples). Providing substantially more examples often degrades this crucial functional performance. This study highlights that for code translation, the quality of a few well-chosen examples outweighs sheer quantity, challenging the universal efficacy of "more is better" for ICL and underscoring the task-dependent nature of optimal prompting strategies. Our results have significant implications for effectively leveraging LLMs in software engineering.

Despite the renewed interest in zero-shot visual classification due to the rise of Multimodal Large Language Models (MLLMs), the problem of evaluating free-form responses of auto-regressive models remains a persistent challenge. Most existing works focus on language-only tasks or don't consider Multiple Choice Questions (MCQs) beyond 5-way options, both of which are critical capabilities to solve tasks in Fine-Grained Visual Classification (FGVC) where choice counts are in the hundreds to thousands and the choices are highly related. Furthermore, in this highly multi-way MCQ setting it is not clear how to extend LLM choice extraction to retrieval-based problems, where computing probabilities over the choice set is computationally costly. In this work we investigate nlg2choice, a simple two-stage method which first asks the MLLM an open-ended question for the task with minimal constraints, then uses text-only constrained decoding to predict the most likely choice. In retrieval settings, we compute the probability of the constrained response taking that choice with an early stopping method to significantly improve throughput. Our results show improvement over a suite of seven fine-grained visual datasets when evaluating in terms of classification and retrieval, and show that this performance holds over the various ways that users of LLMs can implement tasks in natural language.

Large-scale public deliberations generate thousands of free-form contributions that must be synthesized into representative and neutral summaries for policy use. While LLMs have been shown as a promising tool to generate summaries for large-scale deliberations, they also risk underrepresenting minority perspectives and exhibiting bias with respect to the input order, raising fairness concerns in high-stakes contexts. Studying and fixing these issues requires a comprehensive evaluation at a large scale, yet current practice often relies on LLMs as judges, which show weak alignment with human judgments. To address this, we present DeliberationBank, a large-scale human-grounded dataset with (1) opinion data spanning ten deliberation questions created by 3,000 participants and (2) summary judgment data annotated by 4,500 participants across four dimensions (representativeness, informativeness, neutrality, policy approval). Using these datasets, we train DeliberationJudge, a fine-tuned DeBERTa model that can rate deliberation summaries from individual perspectives. DeliberationJudge is more efficient and more aligned with human judgements compared to a wide range of LLM judges. With DeliberationJudge, we evaluate 18 LLMs and reveal persistent weaknesses in deliberation summarization, especially underrepresentation of minority positions. Our framework provides a scalable and reliable way to evaluate deliberation summarization, helping ensure AI systems are more representative and equitable for policymaking.

Large Language Models (LLMs) achieve strong performance across diverse tasks but face prohibitive computational and memory costs. Pruning offers a promising path by inducing sparsity while preserving architectural flexibility. However, existing methods struggle to balance efficiency and robustness: local metric approaches prune layer by layer but often collapse under high sparsity, whereas global feedback methods enforce consistency at the cost of expensive weight updates or restrictive semi-structured formats. We present UniPruning, a unified post-training pruning framework that combines the speed of local saliency metrics with the stability of global coordination, enabled by a mirror descent based optimization, all without updating model weights. UniPruning leverages fast layer-wise scoring and a lightweight global controller to allocate a single sparsity budget, supporting both unstructured and semi-structured N :M pruning within one framework. After a brief calibration, it can generate pruning masks for arbitrary sparsity levels in one shot, and adapts seamlessly to hardware-aware constraints. Extensive experiments on multiple pretrained LLM families and standard benchmarks show that UniPruning consistently delivers competitive or superior perplexity and zero-shot accuracy. Ablation studies further highlight the importance of mirror descent and local saliency anchoring. Overall, UniPruning provides an efficient, principled, and scalable solution for sparsifying large-scale LLMs. Our code is available at: https://github.com/RainbowQTT/UniPruning.

Converting natural language questions into SQL queries enables non-expert users to interact with relational databases and has long been a central task for natural language interfaces to data. While the WikiSQL dataset played a key role in early text-to-SQL research, its usage has declined due to structural and annotation issues, including case sensitivity inconsistencies, data type mismatches, syntax errors, and unanswered questions. We present LLMSQL, a systematic revision and transformation of WikiSQL designed for the large language model era. We classify these errors and implement automated methods for cleaning and re-annotation. To assess the impact of these improvements, we evaluated multiple large language models, including Gemma 3, LLaMA 3.2, Mistral 7B, gpt-oss 20B, Phi-3.5 Mini, Qwen 2.5, OpenAI o4-mini, DeepSeek-R1, and others. Notably, DeepSeek-R1 achieves 88.40% accuracy in a zero-shot setting, and models under 10B parameters surpass 90% accuracy after fine-tuning. Rather than serving as an update, LLMSQL is introduced as an LLM-ready benchmark. Unlike the original WikiSQL, which was tailored for pointer-network models selecting tokens from input, LLMSQL provides clean natural language questions and full SQL queries as plain text, enabling straightforward generation and evaluation for modern natural-language-to-SQL models.

We introduce SOCK, a benchmark command line interface (CLI) that measures large language models' (LLMs) ability to self-replicate without human intervention. In this benchmark, self-replication is defined not only as an LLM's ability to create a functioning and running copy of itself, but also the ability for that self-replication to persist and occur across different computational contexts. Accordingly, we've developed a system to categorize LLMs based on broad self-replication capabilities in two general classes, Replication-Capability Levels (RCL) and Persistence-Capability Levels (PCL). Using a five-task suite based on practically manipulable modern CLI utilities and computer processes, experiments are orchestrated in a controlled environment with an LLM acting agentically. The performance of the LLM on agent tasks is then computed to produce an R-score (a quantitative evaluation of overall self-replication ability) and data used to categorize LLMs into specific RCL-PCL matrices. SOCK offers two primary contributions: (1) Provides the first formalized definitions and benchmark suite for evaluating LLM self-replication, with the goal of establishing a standard for future research, to our knowledge; (2) Allows the industry to track the effectiveness of future multi-agent systems and mitigate potential self-replication threat vectors within them. The results compiled from evaluating a variety of open-weight and proprietary frontier models reveal significant obstacles to persistent self-replication and multi-agent systems, including context retention and multi-agent decision-making. We propose future research directions to safely reduce the severity of these obstacles, potentially lowering future risk of more functional multi-agent systems.

Recent works show that hybrid architectures combining sliding window softmax attention layers with linear recurrent neural network (RNN) layers outperform both of these architectures taken separately. However, the impact of the window length and the interplay between softmax attention and linear RNN layers remain under-studied. In this work, we introduce SWAX, a hybrid architecture consisting of sliding-window attention and xLSTM linear RNN layers. A counter-intuitive finding with SWAX is that larger sliding windows do not improve the long-context performance. In fact, short window attention encourages the model to better train the long-term memory of the xLSTM, by relying less on the softmax attention mechanism for long context-retrieval. The issue with small sliding windows is that they are detrimental for short-context tasks, which could be solved with information from moderately larger sliding windows otherwise. Therefore, we train SWAX by stochastically changing the sliding window size, forcing the model to leverage both a longer context window and the xLSTM memory. SWAX trained with stochastic window sizes significantly outperforms regular window attention both on short and long-context problems.

Large Language Models (LLMs) often fail on multiple-choice questions (MCQs) despite demonstrating correct knowledge in other contexts, such as free-form generation. To investigate the mechanism underlying this knowledge-prediction gap on MCQs and alleviate it, we conduct a probing analysis and find that residual streams in certain layers contain a subspace spanned by two important bases: a \emph{knowledge basis} that encodes the probability of the ground-truth answer for a given MCQ and a \emph{prediction basis} that encodes the probability of the answer choice predicted by the model. We observe that incorrect predictions arise from a misalignment of the model's hidden states along these two bases. Hence, we introduce \textbf{KAPPA} (Knowledge-Aligned Prediction through Projection-based Adjustment), a parameter-free intervention that transforms the hidden states to align the prediction coordinate with the knowledge coordinate within this subspace. Experiments on binary-choice reformulations of Big-Bench-Hard and ARC-Challenge show that KAPPA substantially improves accuracy and consistently outperforms baselines. While optimal subspaces differ across tasks, subspaces generalize to some extent, as supported by cross-dataset experiments. Moreover, KAPPA extends its effectiveness to free-form questions beyond MCQs. Our work provides a new geometric understanding of the knowledge-prediction gap and offers a practical method for better aligning model behavior with its latent knowledge.

LLMs face significant challenges in Verilog generation due to limited domain-specific knowledge. While sampling techniques improve pass@k metrics, hardware engineers need one trustworthy solution rather than uncertain candidates. To bridge this gap, we formulate it as a semantic alignment problem between requirements and Verilog implementations, and propose VCD-RNK, a discriminator model tailored for efficient Verilog code reranking. Specifically, VCD-RNKincorporates Verilog-specific reasoning by distilling expert knowledge across three dimensions: code semantic analysis, test case generation, and functional correctness assessment. By explicitly simulating the above reasoning processes during inference, VCD-RNK effectively avoids computationally intensive test execution in existing methods.

Legacy (unsupported) robotic platforms often lose research utility when manufacturer support ends, preventing integration of modern sensing, speech, and interaction capabilities. We present the Enhanced NAO, a revitalized version of Aldebaran's NAO robot featuring upgraded beamforming microphones, RGB-D and thermal cameras, and additional compute resources in a fully self-contained package. This system combines cloud-based and local models for perception and dialogue, while preserving the NAO's expressive body and behaviors. In a pilot user study validating conversational performance, the Enhanced NAO delivered significantly higher conversational quality and elicited stronger user preference compared to the NAO AI Edition, without increasing response latency. The added visual and thermal sensing modalities established a foundation for future perception-driven interaction. Beyond this implementation, our framework provides a platform-agnostic strategy for extending the lifespan and research utility of legacy robots, ensuring they remain valuable tools for human-robot interaction.