❌

Normal view

Received — 8 April 2026 ⏭

Individual and Combined Effects of English as a Second Language and Typos on LLM Performance

✇cs.AI, q-bio.NC updates on arXiv.org
By: Serena Liu · Yutong Yang · Prisha Sheth · Weixuan Dong · Mingjiao Diao · Xinru Zhu · Nikhil Banga · Oscar Melendez · Arnav Sharma · Minda Zhao · Marina Lin · Mengyu Wang
7 April 2026 at 12:00
arXiv:2604.04723v1 Announce Type: cross Abstract: Large language models (LLMs) are used globally, and because much of their training data is in English, they typically perform best on English inputs. As a result, many non-native English speakers interact with them in English as a second language (ESL), and these inputs often contain typographical errors. Prior work has largely studied the effects of ESL variation and typographical errors separately, even though they often co-occur in real-world use. In this study, we use the Trans-EnV framework to transform standard English inputs into eight ESL variants and apply MulTypo to inject typos at three levels: low, moderate, and severe. We find that combining ESL variation and typos generally leads to larger performance drops than either factor alone, though the combined effect is not simply additive. This pattern is clearest on closed-ended tasks, where performance degradation can be characterized more consistently across ESL variants and typo levels, while results on open-ended tasks are more mixed. Overall, these findings suggest that evaluations on clean standard English may overestimate real-world model performance, and that evaluating ESL variation and typographical errors in isolation does not fully capture model behavior in realistic settings.
Received — 26 March 2026 ⏭

Genomic history of early dogs in Europe

✇Nature - Issue - nature.com science feeds
By: Anders Bergström · Anja Furtwängler · Sarah Johnston · Erika Rosengren · Abagail Breidenstein · Thomas Booth · Jesse B. McCabe · Jessica Peto · Mia Williams · Monica Kelly · Frankie Tait · Chris Baumann · Rita Radzeviciute · Christopher Barrington · Kyriaki Anastasiadou · Alexandre Gilardet · Isabelle Glocke · Mattias Sherman · Anastasia Brativnyk · Alexander Herbig · Kay Prüfer · Saskia Pfrengle · Joscha Gretzinger · Tatiana R. Feuerborn · Ella Reiter · Anna Linderholm · Sophy Charlton · Fernando Racimo · Lea Mikkola · Hugo Anderson-Whymark · Douglas Baird · Anne Birgitte Gotfredsen · Hervé Bocherens · Anne Bridault · Rainer Brocke · Dorothée G. Drucker · Andrew S. Fairbairn · Laurent Frantz · Boris Gasparyan · Liane Giemsch · Mietje Germonpré · Luc Janssens · Andrew W. Kandel · Kurt Kjær · Martina Lázničková-Galetová · Daniel Loponte · Ola Magnell · Louise Martin · Susanne C. Münzel · Gökhan Mustafaoğlu · Bjørnar Måge · Angela Perri · Franziska Pfenninger · Martina Roblíčková · Annelise Roman-Binois · Özlem Sarıtaş · Katharina Schäppi · J.
25 March 2026 at 08:00

Nature, Published online: 25 March 2026; doi:10.1038/s41586-026-10112-7

Genome-wide analysis shows European dogs existed by 14,200 years ago, were already genetically distinct, received less Neolithic Southwest Asian admixture than humans did and contributed substantially to later European dogs.

Dogs were widely distributed across western Eurasia during the Palaeolithic

✇Nature - Issue - nature.com science feeds
By: William A. Marsh · Lachie Scarsbrook · Eren Yüncü · Lizzie Hodgson · Audrey T. Lin · Maria De Iorio · Olaf Thalmann · Mark G. Thomas · Mahaut Goor · Anders Bergström · Angela Noseda · Sarieh Amiri · Fereidoun Biglari · Dušan Borić · Katia Bougiouri · Alberto Carmagnini · Maddalena Giannì · Tom Higham · Ophelie Lebrasseur · Anna Linderholm · Marcello A. Mannino · Caroline Middleton · Gökhan Mustafaoğlu · Angela Perri · Joris Peters · Mike Richards · Özlem Sarıtaş · Pontus Skoglund · Rhiannon E. Stevens · Chris Stringer · Kristina Tabbada · Helen M. Talbot · Laura G. Van der Sluis · Silvia M. Bello · Vesna Dimitrijevic · Louise Martin · Marjan Mashkour · Simon A. Parfitt · Sonja Vukovic · Selina Brace · Oliver E. Craig · Douglas Baird · Sophy Charlton · Greger Larson · Ian Barnes · Laurent A. F. Frantz
25 March 2026 at 08:00

Nature, Published online: 25 March 2026; doi:10.1038/s41586-026-10170-x

Analysis of nuclear and mitochondrial genomes from archaeological canid remains found across Europe and Anatolia shows that a genetically homogeneous dog population was already widely distributed across the region by 15,000 years ago.

A Review of the Role of Zeqi Decoction in the Treatment of Non-Small Cell Lung Cancer

✇Pulmonary nodule
By: Lan Zhang · Yanhui Jia · Lei Song · Yiman Li · Yawei Liu · Yan Qin · Tao Wang · Na Li
18 March 2026 at 18:00

J Multidiscip Healthc. 2026 Mar 11;19:584071. doi: 10.2147/JMDH.S584071. eCollection 2026.

ABSTRACT

Non-small cell lung cancer (NSCLC) is one of the malignant tumors with the highest incidence and mortality rates. Zeqi Decoction has the functions of "promoting diuresis and reducing swelling, resolving phlegm and dispersing nodules", embodying the unique approach of traditional Chinese medicine in treating lung cancer by "strengthening the body's resistance and eliminating pathogenic factors". Modern research shows that Zeqi Decoction exerts anti-NSCLC effects through multiple pathways and targets. In terms of the material basis of its efficacy, its active ingredients (such as diterpene esters and flavonoids contained in Zeqi) have the ability to directly inhibit the proliferation, invasion and migration of tumor cells and induce apoptosis. In terms of the mechanism of action, basic experiments have revealed that Zeqi Decoction can down-regulate the S100A9/STAT3 signaling pathway, inhibit the immunosuppressive activity of myelium-derived suppressor cells (MDSCs), reshape the tumor microenvironment, thereby enhancing the cytotoxic function of CD8⁺T cells, and can also regulate the EGFR/PI3K/Akt pathway to affect PD-L1 expression. Intervene in tumor immune escape; In terms of clinical transformation, the combination of Zexi Decoction with chemotherapy and targeted therapy can improve patients' symptoms such as cough and pleural effusion, prolong progression-free survival, and alleviate the toxic and side effects of Western medical treatment. In addition, Zexi Decoction also shows potential value in reversing drug resistance such as gemcitabine. At present, there are still problems such as the lack of standardized protocols and unclear molecular mechanisms in the research. In the future, it is necessary to combine new technologies such as network pharmacology and multi-omics analysis to deepen the research on the pharmacological material basis, dose-effect relationship and evidence-based medicine of Zeqi Decoction, so as to promote the clinical application and transformation of the combination of traditional Chinese and Western medicine in the treatment of NSCLC.

PMID:41847115 | PMC:PMC12991379 | DOI:10.2147/JMDH.S584071

Received — 5 March 2026 ⏭

R1-Code-Interpreter: LLMs Reason with Code via Supervised and Multi-stage Reinforcement Learning

✇cs.AI, q-bio.NC updates on arXiv.org
By: Yongchao Chen · Yueying Liu · Junwei Zhou · Yilun Hao · Jingquan Wang · Yang Zhang · Na Li · Chuchu Fan
5 March 2026 at 13:00
arXiv:2505.21668v3 Announce Type: replace Abstract: Practical guidance on training Large Language Models (LLMs) to leverage Code Interpreter across diverse tasks remains lacking. We present R1-Code-Interpreter, an extension of a text-only LLM trained via multi-turn supervised fine-tuning (SFT) and reinforcement learning (RL) to autonomously generate multiple code queries during step-by-step reasoning. Unlike prior RL + tool-use efforts focused on narrow domains such as math or retrieval, we curate 144 diverse reasoning and planning tasks and show that training a general-purpose Code Interpreter across them presents significant challenges due to task heterogeneity and scarcity of effective samples. To address this, we introduce a multi-stage curriculum learning approach that partitions training samples by measured improvement potential. The RL training prioritizes samples with higher potential and gradually shifts to lower-potential ones, increasing the average RL gains from merely +3.4% to +9.3% across Qwen-2.5 models (3/7/14B). Our final model, R1-CI-14B, improves average accuracy on the 37 test tasks from 44.1% to 72.4%, outperforming text-only GPT-4o (58.6%) and GPT-4o with Code Interpreter (70.9%). Notably, R1-CI-14B also exhibits emergent self-checking behavior through code generation. Datasets, Codes, and Models are available at https://github.com/yongchao98/R1-Code-Interpreter and https://huggingface.co/yongchao98.
Received — 25 February 2026 ⏭

Closing the Gap Between Text and Speech Understanding in LLMs

✇cs.AI, q-bio.NC updates on arXiv.org
By: Santiago Cuervo · Skyler Seto · Maureen de Seyssel · Richard He Bai · Zijin Gu · Tatiana Likhomanenko · Navdeep Jaitly · Zakaria Aldeneh
24 February 2026 at 13:00
arXiv:2510.13632v2 Announce Type: replace-cross Abstract: Large Language Models (LLMs) can be adapted to extend their text capabilities to speech inputs. However, these speech-adapted LLMs consistently underperform their text-based counterparts--and even cascaded pipelines--on language understanding tasks. We term this shortfall the text-speech understanding gap: the performance drop observed when a speech-adapted LLM processes spoken inputs relative to when the original text-based LLM processes the equivalent text. Recent approaches to narrowing this gap either rely on large-scale speech synthesis of text corpora, which is costly and heavily dependent on synthetic data, or on large-scale proprietary speech datasets, which are not reproducible. As a result, there remains a need for more data-efficient alternatives for closing the text-speech understanding gap. In this work, we analyze the gap as driven by two factors: (i) forgetting of text capabilities during adaptation, and (ii) cross-modal misalignment between speech and text. Based on this analysis, we introduce SALAD--Sample-efficient Alignment with Learning through Active selection and cross-modal Distillation--which combines cross-modal distillation with targeted synthetic data to improve alignment while mitigating forgetting. Applied to 3B and 7B LLMs, SALAD achieves competitive performance with a strong open-weight model across broad-domain benchmarks in knowledge, language understanding, and reasoning, while training on over an order of magnitude less speech data from public corpora.

EBPO: Empirical Bayes Shrinkage for Stabilizing Group-Relative Policy Optimization

✇cs.AI, q-bio.NC updates on arXiv.org
By: Kevin Han · Yuhang Zhou · Mingze Gao · Gedi Zhou · Serena Li · Abhishek Kumar · Xiangjun Fan · Weiwei Li · Lizhu Zhang
24 February 2026 at 13:00
arXiv:2602.05165v3 Announce Type: replace-cross Abstract: Reinforcement Learning with Verifiable Rewards (RLVR) has proven effective for enhancing the reasoning capabilities of Large Language Models (LLMs). However, dominant approaches like Group Relative Policy Optimization (GRPO) face critical stability challenges: they suffer from high estimator variance under computational constraints (small group sizes) and vanishing gradient signals in saturated failure regimes where all responses yield identical zero rewards. To address this, we propose Empirical Bayes Policy Optimization (EBPO), a novel framework that regularizes local group-based baselines by borrowing strength from the policy's accumulated global statistics. Instead of estimating baselines in isolation, EBPO employs a shrinkage estimator that dynamically balances local group statistics with a global prior updated via Welford's online algorithm. Theoretically, we demonstrate that EBPO guarantees strictly lower Mean Squared Error (MSE), bounded entropy decay, and non-vanishing penalty signals in failure scenarios compared to GRPO. Empirically, EBPO consistently outperforms GRPO and other established baselines across diverse benchmarks, including AIME and OlympiadBench. Notably, EBPO exhibits superior training stability, achieving high-performance gains even with small group sizes, and benefits significantly from difficulty-stratified curriculum learning.
Received — 17 February 2026 ⏭

What Do We Mean by 'Pilot Study': Early Findings from a Meta-Review of Pilot Study Reporting at CHI

✇cs.AI, q-bio.NC updates on arXiv.org
By: Belu Ticona · Amna Liaqat · Antonios Anastasopoulos
17 February 2026 at 13:00
arXiv:2602.13488v1 Announce Type: cross Abstract: Pilot studies (PS) are ubiquitous in HCI research. CHI papers routinely reference 'pilot studies', 'pilot tests', or 'preliminary studies' to justify design decisions, verify procedures, or motivate methodological choices. Yet despite their frequency, the role of pilot studies in HCI remains conceptually vague and empirically underexamined. Unlike fields such as medicine, nursing, and education, where pilot and feasibility studies have well-established definitions, guidelines, reporting standards and even a dedicated research journal, the CHI community lacks a shared understanding of what constitutes a pilot study, why they are conducted, and how they should be reported. Many papers reference pilots 'in passing', without details about design, outcomes, or how the pilot informed the main study. This variability suggests a methodological blind spot in our community.
There are no more articles
❌