Tasks on complex systems require high-precision numerical computation to support decisions, but current large language models (LLMs) cannot integrate such computations as an intrinsic and interpretable capability with existing architectures. Multi-agent approaches can leverage external experts, but inevitably introduce communication overhead and suffer from inefficiency caused by limited scalability. To this end, we propose Physically-isolated Experts Routing Network (PiERN), an architecture for integrating computation and reasoning. Instead of the tool-use workflows or function-calling, PiERN endogenously integrates computational capabilities into neural networks after separately training experts, a text-to-computation module, and a router. At inference, the router directs computation and reasoning at the token level, thereby enabling iterative alternation within a single chain of thought. We evaluate PiERN on representative linear and nonlinear computation-reasoning tasks against LLM finetuning and the multi-agent system approaches. Results show that the PiERN architecture achieves not only higher accuracy than directly finetuning LLMs but also significant improvements in response latency, token usage, and GPU energy consumption compared with mainstream multi-agent approaches. PiERN offers an efficient, interpretable, and scalable paradigm for interfacing language models with scientific systems.

Creators struggle to edit long-form, narrative-rich videos not because of UI complexity, but due to the cognitive demands of searching, storyboarding, and sequencing hours of footage. Existing transcript- or embedding-based methods fall short for creative workflows, as models struggle to track characters, infer motivations, and connect dispersed events. We present a prompt-driven, modular editing system that helps creators restructure multi-hour content through free-form prompts rather than timelines. At its core is a semantic indexing pipeline that builds a global narrative via temporal segmentation, guided memory compression, and cross-granularity fusion, producing interpretable traces of plot, dialogue, emotion, and context. Users receive cinematic edits while optionally refining transparent intermediate outputs. Evaluated on 400+ videos with expert ratings, QA, and preference studies, our system scales prompt-driven editing, preserves narrative coherence, and balances automation with creator control.

Large language models (LLMs) benefit from test-time scaling but are often hampered by high inference latency. Speculative decoding is a natural way to accelerate the scaling process; however, scaling along both the parallel and sequential dimensions poses significant challenges, including substantial memory-bound execution and synchronization overhead. We introduce ATTS (Asynchronous Test-Time Scaling), a statistically guaranteed adaptive scaling framework that follows the hypothesis testing process to address these challenges. By revisiting arithmetic intensity, ATTS identifies synchronization as the primary bottleneck. It enables asynchronous inference through online calibration and proposes an ordinal classification algorithm that supports a three-stage rejection sampling pipeline, scaling along both the sequential and parallel axes. Across experiments on the MA TH, AMC23, AIME24, and AIME25 datasets and across multiple draft-target model families, we show that ATTS delivers up to 56.7x speedup in test-time scaling and a 4.14x throughput improvement, while maintaining accurate control of the rejection rate, reducing latency and memory overhead, and incurring no accuracy loss. By scaling both in parallel and sequential dimensions, we enable the 1.5B/70B draft/target model combination to achieve the performance of the state-of-the-art reasoning model o3-mini (high) on the AIME dataset. With the rapid advances in large language models (LLMs), attention is increasingly turning to reasoning models (Guo et al., 2025; Muennighoff et al., 2025; McCoy et al., 2024; Shao et al., 2024)--systems that transcend next-token prediction in order to emulate human-like reasoning behaviors.

This position paper examines how large language models (LLMs) can support thematic analysis of unstructured clinical transcripts, a widely used but resource-intensive method for uncovering patterns in patient and provider narratives. We conducted a systematic review of recent studies applying LLMs to thematic analysis, complemented by an interview with a practicing clinician. Our findings reveal that current approaches remain fragmented across multiple dimensions including types of thematic analysis, datasets, prompting strategies and models used, most notably in evaluation. Existing evaluation methods vary widely (from qualitative expert review to automatic similarity metrics), hindering progress and preventing meaningful benchmarking across studies. We argue that establishing standardized evaluation practices is critical for advancing the field. To this end, we propose an evaluation framework centered on three dimensions: validity, reliability, and interpretability.

The most widely used large language models in the social sciences (such as BERT, and its derivatives, e.g. RoBERTa) have a limitation on the input text length that they can process to produce predictions. This is a particularly pressing issue for some classification tasks, where the aim is to handle long input texts. One such area deals with laws and draft laws (bills), which can have a length of multiple hundred pages and, therefore, are not particularly amenable for processing with models that can only handle e.g. 512 tokens. In this paper, we show results from experiments covering 5 languages with XLM-RoBERTa, Longformer, GPT-3.5, GPT-4 models for the multiclass classification task of the Comparative Agendas Project, which has a codebook of 21 policy topic labels from education to health care. Results show no particular advantage for the Longformer model, pre-trained specifically for the purposes of handling long inputs. The comparison between the GPT variants and the best-performing open model yielded an edge for the latter. An analysis of class-level factors points to the importance of support and substance overlaps between specific categories when it comes to performance on long text inputs.

Does continued scaling of large language models (LLMs) yield diminishing returns? In this work, we show that short-task benchmarks may give an illusion of slowing progress, as even marginal gains in single-step accuracy can compound into exponential improvements in the length of tasks a model can successfully complete. Then, we argue that failures of LLMs when simple tasks are made longer arise from mistakes in execution, rather than an inability to reason. So, we propose isolating execution capability, by explicitly providing the knowledge and plan needed to solve a long-horizon task. First, we find that larger models can correctly execute significantly more turns even when small models have near-perfect single-turn accuracy. We then observe that the per-step accuracy of models degrades as the number of steps increases. This is not just due to long-context limitations -- curiously, we observe a self-conditioning effect -- models become more likely to make mistakes when the context contains their errors from prior turns. Self-conditioning does not reduce by just scaling the model size. But, we find that thinking mitigates self-conditioning, and also enables execution of much longer tasks in a single turn. We conclude by benchmarking frontier thinking models on the length of tasks they can execute in a single turn. Overall, by focusing on the ability to execute, we hope to reconcile debates on how LLMs can solve complex reasoning problems yet fail at simple tasks when made longer, and highlight the massive benefits of scaling model size and sequential test-time compute for long-horizon tasks.

Language use is shaped by pragmatics -- i.e., reasoning about communicative goals and norms in context. As language models (LMs) are increasingly used as conversational agents, it becomes ever more important to understand their pragmatic reasoning abilities. We propose an evaluation framework derived from Wavelength, a popular communication game where a speaker and a listener communicate about a broad range of concepts in a granular manner. We study a range of LMs on both language comprehension and language production using direct and Chain-of-Thought (CoT) prompting, and further explore a Rational Speech Act (RSA) approach to incorporating Bayesian pragmatic reasoning into LM inference. We find that state-of-the-art LMs, but not smaller ones, achieve strong performance on language comprehension, obtaining similar-to-human accuracy and exhibiting high correlations with human judgments even without CoT prompting or RSA. On language production, CoT can outperform direct prompting, and using RSA provides significant improvements over both approaches. Our study helps identify the strengths and limitations in LMs' pragmatic reasoning abilities and demonstrates the potential for improving them with RSA, opening up future avenues for understanding conceptual representation, language understanding, and social reasoning in LMs and humans.

As large language models (LLMs) expand multilingual capabilities, questions remain about the equity of their performance across languages. While many communities stand to benefit from AI systems, the dominance of English in training data risks disadvantaging non-English speakers. To test the hypothesis that such data disparities may affect model performance, this study compares two monolingual BERT models: one trained and tested entirely on Swahili data, and another on comparable English news data. To simulate how multilingual LLMs process non-English queries through internal translation and abstraction, we translated the Swahili news data into English and evaluated it using the English-trained model. This approach tests the hypothesis by evaluating whether translating Swahili inputs for evaluation on an English model yields better or worse performance compared to training and testing a model entirely in Swahili, thus isolating the effect of language consistency versus cross-lingual abstraction. The results prove that, despite high-quality translation, the native Swahili-trained model performed better than the Swahili-to-English translated model, producing nearly four times fewer errors: 0.36% vs. 1.47% respectively. This gap suggests that translation alone does not bridge representational differences between languages and that models trained in one language may struggle to accurately interpret translated inputs due to imperfect internal knowledge representation, suggesting that native-language training remains important for reliable outcomes. In educational and informational contexts, even small performance gaps may compound inequality. Future research should focus on addressing broader dataset development for underrepresented languages and renewed attention to multilingual model evaluation, ensuring the reinforcing effect of global AI deployment on existing digital divides is reduced.

Reinforcement Learning (RL) has proven highly effective at enhancing the complex reasoning abilities of Large Language Models (LLMs), yet underlying mechanisms driving this success remain largely opaque. Our analysis reveals that puzzling phenomena like ``aha moments", ``length-scaling'' and entropy dynamics are not disparate occurrences but hallmarks of an emergent reasoning hierarchy, akin to the separation of high-level strategic planning from low-level procedural execution in human cognition. We uncover a compelling two-phase dynamic: initially, a model is constrained by procedural correctness and must improve its low-level skills. The learning bottleneck then decisively shifts, with performance gains being driven by the exploration and mastery of high-level strategic planning. This insight exposes a core inefficiency in prevailing RL algorithms like GRPO, which apply optimization pressure agnostically and dilute the learning signal across all tokens. To address this, we propose Hierarchy-Aware Credit Assignment (HICRA), an algorithm that concentrates optimization efforts on high-impact planning tokens. Our extensive experiments validate that HICRA significantly outperforms strong baselines, and offer deep insights into how reasoning advances through the lens of strategic exploration.

Do LLMs genuinely incorporate external definitions, or do they primarily rely on their parametric knowledge? To address these questions, we conduct controlled experiments across multiple explanation benchmark datasets (general and domain-specific) and label definition conditions, including expert-curated, LLM-generated, perturbed, and swapped definitions. Our results reveal that while explicit label definitions can enhance accuracy and explainability, their integration into an LLM's task-solving processes is neither guaranteed nor consistent, suggesting reliance on internalized representations in many cases. Models often default to their internal representations, particularly in general tasks, whereas domain-specific tasks benefit more from explicit definitions. These findings underscore the need for a deeper understanding of how LLMs process external knowledge alongside their pre-existing capabilities.