跳到正文
研踪PaperTrail

论文库

检索计算机领域论文,按研究方向、作者或关键词筛选。

部分采集异常
正在读取账号与研究记录…

全站更新概况

+1112新增论文
+875其中 AI 相关
+133已有论文修订
+97新增 AI 导读

arXiv 最近完成扫描:2026-11-20 05:42 · 全站最近入库:2026-11-19 19:09各信源进度 →

统计口径与数据新鲜度

统计 2026-11-19 05:48 — 2026-11-20 05:48(北京时间) · 点击数字查看对应简报;AI 相关包含在新增论文内。

最近完成检查:2026-11-20 05:42最近发现新增版本:2026-11-19 19:09库内论文最新更新时间:2026-11-19 02:59

每分钟刷新已保存的数据。空检查不代表采集异常;新收录可能包含历史论文。AI 导读按新增版本计数。

27,982 已收录论文23,249 AI 相关8,559 有修订112 HF 社区发现

浏览论文库

时间范围跟随排序:版本更新、本站收录或首次发表。最新收录可能包含历史论文。部分信源异常,查看状态 →

找到 21 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-20 05:40(北京时间)
新论文 · v1arXiv:2611.24650

Self-Rewarded Reinforcement Learning for Better-Structured Medical Explanations in Small Language Models

Prakhar Mishra, Bhavya Yajush Awasthi, Tanuja, Nishant

原文摘要

Small language models (SLMs) can run on commodity hardware and support privacy-preserving, on-premises clinical deployment, but their medical explanations are often short and poorly structured compared with large language models. Recent supervised approaches such as Meerkat transfer chain-of-thought reasoning into 7B-8B backbones, yet their training objective does not directly target the style or organization of the generated rationales. Building on the instruction-tuned Mistral-7B backbone, we investigate whether a lightweight reinforcement learning (RL) step can improve explanation structure without human preference labels.We apply Proximal Policy Optimization (PPO) with LoRA adaptation to 2,000 Meerkat-derived medical prompts, guided by a simple self-supervised reward that favors longer, lexically diverse, and well-punctuated responses while penalizing very short ones. Training for about 170 PPO steps on a single NVIDIA A100 GPU yields a stable increase in the average reward. Quantitative and qualitative comparisons show that the RL-tuned model produces longer, more organized, multi-point explanations while preserving the underlying medical content.We discuss the promise and the limitations of self-rewarded RL for shaping the style of medical explanations.

cs.AI · cs.CL

新论文 · v1arXiv:2611.20929

Transferable Neural CBFs via Generalized Closure

Nishant Jayesh Bhave, Parthib Khound

原文摘要

Classical control barrier functions (CBFs) provide safety constraints for controllers, but designing them by hand is difficult for underactuated vehicles and the resulting safe sets can be overly conservative. Neural CBFs can learn more informative barriers from data, yet a learned barrier is usually verified only for one fixed obstacle configuration. We study whether one neural barrier can be reused across environments by expressing the state relative to each obstacle. The key gap is that these relative coordinates may discard information needed to predict their own evolution. We give a generalized closure condition that represents this missing information as a residual and a lifting theorem that transfers a verified relative-coordinate barrier to every obstacle in a family. A rank test also identifies when symmetry alone cannot justify the transfer. We validate the conditions on a double integrator, a planar aircraft, and a 12-state quadrotor. Coordinates that fail the test lead to obstacle penetration or to a barrier condition that fails near hover, whereas body-frame coordinates close exactly for the quadrotor. The resulting barrier transfers to new maps without retraining and is effective in the reported deployment and fast-start experiments.

cs.RO · eess.SY

新论文 · v1arXiv:2611.18024

JaRS: Contrast-Free Localization of Memorized Regions in Diffusion Models via Jacobian Row Sums

Saisuchith Mahajan, Aditey Nandan, Rohit Singh, Nishanth Shetty, Chandra Sekhar Seelamantula

原文摘要

Diffusion models sometimes memorize and reproduce regions of their training images, raising privacy and copyright concerns. We aim to localize these regions. Existing contrast-based methods use the diagonal of the denoiser Jacobian, which reflects the posterior variance of each pixel. However, a small diagonal entry is ambiguous: it signals either memorization or averaging over many neighboring pixels in smooth or predictable regions. To resolve this ambiguity, these methods subtract a reference map, which requires a second evaluation with a null prompt or an earlier checkpoint. Unconditional models have no null prompt, and earlier checkpoints are unavailable in general. We show that Jacobian row sums (JaRS), unlike the diagonal, separate memorization from local averaging without a reference map. Pairing JaRS with an image-level detection score that uses the same statistic at an early denoising step yields JaRS+. Across Stable Diffusion v1.4 and v2.1, Realistic Vision v5.1, and an unconditional diffusion Transformer (DiT), JaRS+ outperforms contrast-based localization methods.

cs.CV · cs.LG

新论文 · v1arXiv:2611.14477

FERN:面向Text-to-SQL的跨基准表检索,采用冻结检索器与学习排序

FERN: Cross-Benchmark Table Retrieval for Text-to-SQL with Frozen Retrievers and Learned Ranking

Arjun Rao, Nishant Gopinath

AI 速读 基于摘要

FERN 将三个冻结检索器融合为候选池,并用 LightGBM 排序器重排,在跨 BIRD、Spider、Fiben 的表检索中与 RASL 的已发表召回率进行比较。

cs.IR · cs.CL

查看原文摘要

Text-to-SQL systems over large pooled catalogs must first retrieve the tables a question needs. We compare against the published table recall of RASL, which uses Claude 3.5 models to extract keywords and select tables, on pooled BIRD, Spider, and Fiben catalogs. FERN (Feature-Engineered Ranking with No generative LLM) instead fuses three frozen retrievers into 100 candidates, describes each with up to 120 features, and reorders them with LightGBM rankers fitted once on BIRD's training questions. On Spider, where it was not fitted, its candidate pool alone reaches RASL's published recall at fifteen tables (98.0), and the ranker raises the share of questions with every gold table in the top five (complete@5) from 83.9% to 95.6%. At five tables, FERN's recall is within one point of RASL's on Spider (96.7 vs 97.0) and on BIRD with its evidence hints (96.6 vs 97.5), and 8.2 points below on BIRD without them (89.3). On 121 real BEAVER questions, without BEAVER fitting, its recall at fifteen (82.1) falls between two BEAVER-trained retrievers (80.8, 84.5), but its complete@15 is below both (49.6 vs 54.5 and 62.0). On Fiben, it falls far short at fifteen tables (54.2 vs 69.2, different question sets), although its recall at fifty is numerically close to RASL's question-only retrieval result (80.7 vs 79.4). On BIRD dev with evidence, non-retrieval features raise complete@5 by 7.7 points over retrieval signals alone. On BIRD dev, median latency is 208 ms per question on one A100, without hosted model calls.

新论文 · v1arXiv:2611.12731

NIFT-SQL: Customer-Supervised Table Scoring for Enterprise Text-to-SQL Retrieval

Arjun Rao, Nishant Gopinath

原文摘要

Enterprise text-to-SQL starts with finding the tables a question needs, which is hard when a schema spans several databases and hundreds of cryptic tables. We ask whether models trained separately on each customer's example questions, reading only table names, column names, and declared keys, improve table retrieval. NIFT-SQL (Necessity-Informed Fusion for Text-to-SQL) fuses hybrid search over a retriever fine-tuned on the customer's questions with a fine-tuned 560M cross-encoder that scores every table for whether the query needs it, through a logistic regression fitted on held-out questions. On BEAVER, LiveSQLBench, and LiveSQLBench-Large (244 to 971 tables; four splits, three trainings), NIFT-SQL places every reference table in the top 15 for 69.2%, 89.5%, and 85.6% of test questions, against 30.4%, 41.0%, and 21.8% for hybrid search and 61.1%, 81.7%, and 70.4% for the fine-tuned retriever with the same labels. With off-the-shelf search, the cross-encoder without fine-tuning is 21 points lower. A zero-shot 8B reranker needs no labels and, routed by database, shows no resolved complete@15 difference on two collections; adding its scores as two fusion features gives a high-accuracy mode that reaches 72.4%, 92.5%, and 88.6%, ahead of the validation-selected zero-shot reranker variant on two collections, with no resolved difference on the third. Rank fusion of the two trained parts, grouped by database, shows no resolved difference from the learned fusion. Scored once on a newer BEAVER release, NIFT-SQL beats search and leads the label-matched retriever at top 8 on the synthetic test, but trails it at top 15 on real questions.

cs.IR

新论文 · v1arXiv:2611.11838

Improving Confidence Estimation in Language Models with Linear Representations of Ambiguity

Themistoklis Nikas, Nishant Subramani, Aaron Mueller

原文摘要

As language models (LMs) are increasingly deployed in downstream applications, accurately estimating their likelihood of success on a given example is critical. However, confidence estimates from LMs are poor predictors of their likelihood of success. Accurate confidence estimation is linked in part to the detection of aleatoric uncertainty; we hypothesize that LMs possess representations of at least some aspects of this uncertainty in the form of representations of the inherent ambiguity of the input. Using difference-in-means on minimal pairs of ambiguous and unambiguous sentences, we derive ambiguity representations in LM activations, and find that steering with them selectively (albeit weakly) modulates LMs' judgments of the input's ambiguity. We then propose DART, a fine-tuning method that encourages updates to lie within the span of the ambiguity directions, and penalizes changes to the LM activations outside of this subspace. DART greatly improves calibration, encodes these improvements into the model parameters, and maintains model accuracy on multiple-choice and generative QA tasks.

cs.CL

新论文 · v1arXiv:2611.07327

Permanent and Transient Representations for Continual Reinforcement Learning

Nishanth Anand, Doina Precup

原文摘要

Continual Reinforcement Learning (CRL) agents struggle to adapt to new situations while retaining prior knowledge, leading to a stability-plasticity trade-off. One approach to solving this problem is by using complementary learning systems: one for slow, long-term learning and another for quick, transient adaptation. Anand & Precup (2023) instantiated this idea using permanent and transient value functions, but used the same features in both systems. Intuitively, long-term learning should rely more on parametric representations for broad generalization, while transient learning would be best served by non-parametric representations for situation-specific adaptation. In this paper, we explore this idea both theoretically and empirically. By introducing a novel non-parametric approximation for estimating the transient value function in complex tasks, we demonstrate that our method enables online learning and outperforms competitive baselines on image-based tasks and Craftax-Classic, underscoring the effectiveness of our system-level decomposition.

cs.LG · cs.AI

新论文 · v1arXiv:2611.05779

Private Vulnerability Testing for User Oversharing

Alexander Boyd, Daniel Gardham, Nishanth Sastry, Suparna De

原文摘要

Users can inadvertently increase their vulnerability to privacy attacks by oversharing personal information. In particular, the combination of individually innocuous attributes can place a user in a sufficiently small population to enable linkage or re-identification attacks. We present a privacy-preserving mechanism that allows users to determine whether the information they disclose places them below a predefined population threshold, without revealing their query or the data of other users. Our setting presents several practical challenges: users contributing data may be offline when queries are made, queries may select arbitrary subsets of attributes, and neither server should learn the queried population or its size. We address these challenges using secret-shared data and queries, private equality testing, and secure threshold computation. Our protocol provides computational privacy against a malicious client and semi-honest, non-colluding servers, while supporting arbitrary queries and dynamic database updates. We define reactive $f$-privacy for a dynamically evolving database and prove that our construction satisfies it, with the sequence of permitted threshold outputs defining the information available to the adversary across an execution. We optimise the protocol using oblivious-transfer extensions and pre-computation, achieving sub-second online query times for 96-bit queries on datasets of up to one million records.

cs.CR

新论文 · v1arXiv:2611.04080

FEIJOA: Task-Driven Grounding and Execution for Mobile Manipulation

Aditya Agarwal, William Shen, Sahit Chintalapudi, Lucy Cai, Nishanth Kumar, Aidan Curtis, Travis Manderson, Tomás Lozano-Pérez 等 9 位作者

原文摘要

A mobile manipulator asked to "put the toys away in the toy box" must decide which objects are toys and which container the user means, then find base poses, grasps, and placements it can execute. Because a scene representation is built before the instruction is known, it may lack the attributes, affordances, or concepts needed to resolve the task. We present FEIJOA, a language-model agent that grounds the instruction at test time. It queries a grounding tool over a posed RGB-D scan, inspects the returned candidates, and refines its queries until they resolve the instruction. An embodiment-aware harness checks each skill's preconditions, samples base poses from which the target is reachable, and verifies the skill's effect from observations before and after execution. On 103 open-ended grounding queries, FEIJOA resolves 94.2% exactly, compared with 68.0% for the best prebuilt representation. Over 50 real-world trials on Spot, it completes 35 tasks; 17 of these successes follow recovery from a failed tool call or verification check. Of the 15 failed trials, 13 were caused by the robot's physical skills, and only 2 by grounding or the agent's reasoning. Project website: https://williamshen-nz.github.io/feijoa/.

cs.RO

版本更新 · v2arXiv:2604.16656

Defragmenting Language Models: An Interpretability-based Approach for Vocabulary Expansion

Maitrey Mehta, Nishant Subramani, Zhichao Xu, Ashim Gupta, Vivek Srikumar

原文摘要

All languages are equal; when it comes to tokenization, some are more equal than others. Tokens are the hidden currency that dictates the cost and latency of contemporary LLM access. However, many languages written in non-Latin scripts observe a poor exchange rate: LLMs take several multiples of tokens to encode the same information in many languages as they do for English. We observe that this issue, known as token overfragmentation, persists in modern open-weight LLMs. The standard remedy is vocabulary expansion that adds target language items missing from the model's vocabulary. We focus on two core decisions in the vocabulary expansion process: What items should we add? and How should we initialize their corresponding input and output embeddings? We comprehensively study and advance the new research direction of interpretability-based vocabulary expansion along these two core decisions, making four key contributions: First, we question the conventional use of frequency-based methods to choose candidate vocabulary items to add (a decision long treated as settled), and show that interpretability-based methods offer a superior performance-token efficiency tradeoff. Second, we strengthen the case for interpretability-based embedding initialization by showing large gains (~20 pts) over baseline initialization methods for several languages written in non-Latin scripts. Third, we identify the phenomenon of "subword detokenization," where models progressively merge fragmented subword tokens into larger subwords across layers. Fourth, we propose FragMend, a method that leverages this phenomenon to push the efficiency ceiling of interpretability-based expansion, and validate its effectiveness.

cs.CL

新论文 · v1arXiv:2610.38095

Nemotron 3.5 Super and Lightning: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Models for Agentic AI

NVIDIA, :, Aastha Jhunjhunwala, Aayushi Ashokkumar Srivastava, Abhibha Gupta, Abhinav Garg, Abhinav Khattar, Achsah Wildish 等 559 位作者

原文摘要

We introduce Nemotron 3.5 Super and Nemotron 3.5 Lightning, open and efficient Mixture-of-Experts Hybrid Mamba-Transformer models designed for reasoning, long-context understanding, and agentic workloads. Super is a 120B-parameter, 12B-active multimodal model, while Lightning is a smaller 30B-parameter, 3B-active text model; both designed for efficient deployment. Both models combine large-scale distillation, long-context training, and native speculative decoding, while Super further incorporates Indic continued pretraining and agentic software-engineering mid-training. We develop a comprehensive post-training pipeline spanning supervised fine-tuning and reinforcement learning (RLVR and RLHF), complemented in Super by specialized teacher training and multi- teacher on-policy distillation (MOPD). This transforms Super from a text-only base model into a multimodal agent capable of reasoning over text, images, video, and interactive tool environments. We further explore test-time scaling through parallel agent execution and self-verification. Across training and deployment, we jointly optimize capability and efficiency. Indic continued pretraining improves Indian-language benchmarks by up to 9.82 percentage points, while agentic mid-training strengthens software-engineering capabilities and robustness across agent harnesses. The post-trained Super model achieves 51.78% on SWE-bench Pro, surpassing Nemotron 3 Ultra despite using approximately 4.6x fewer active parameters. Post-training MTP boosting improves speculative- decoding acceptance length by up to 21.8% under probabilistic draft sampling, while NVFP4 quantization achieves 99.62% median accuracy recovery relative to BF16. Inference optimizations enable up to 4,856 output tokens/s/GPU on GB200 while sustaining at least 50 tokens/s/user.

cs.CL · cs.AI · cs.LG

版本更新 · v3arXiv:2607.25130

How Do Developers Edit AI-Generated Code, and How Well Do LLMs Infer Their Edits? Insights from 53.6K Developer Edits

Jenny T. Liang, Mihika Bairathi, Wayne Chi, Ameet Talwalkar, Nishant Subramani, Valerie Chen

原文摘要

Imperfections in AI-generated code require that software developers modify generations through manual code edits. However, little is known about how developers edit AI-generated code, and to what extent LLMs can infer implicit developer intent from this data, given the lack of high-quality, realistic code editing data. This information is essential for designing and building usable AI-developer interactions. To address this gap, we introduce DECODE (Developer Edits of Code Dataset), a dataset of 53.6K real-world in-IDE code edits of AI-generated code sourced from 1K+ developers. First, we use DECODE to analyze when, why, and how AI-generated code is edited. Developers invest little effort in editing code and abandon AI-generated code quickly, thereby motivating the need for LLMs to understand developer intents through their edits. Therefore, we next use DECODE to understand to what extent LLMs could infer developer edits. Predicting code edits is a challenging task for both frontier and fine-tuned models. Frontier models generate edits that best reflect developer intents, but are prone to over-editing and reverting developer-written edits. Fine-tuning LLMs can improve their ability to predict edits that are more similar to developers' as well as better reflect developer intent, but do not fully bridge the gap to frontier models. Together, our results demonstrate the unique insights on developer and LLM behavior that can be obtained through studying realistic code edit data to promote more better developer-AI interactions.

cs.SE · cs.AI · cs.HC · cs.LG

版本更新 · v3arXiv:2610.07853

Lost in the bf16 Cast: Exporting Ternary Language Models Can Revert Most Low-Learning-Rate Code Changes

Avichal Sahai, Nishant Raj, Animesh Srivastava

原文摘要

Ternary language models such as BitNet b1.58, Falcon-E and BitCPM are fine-tuned through higher-precision latents and deployed as ternary codes by an export that, in the labs' documented pipelines, first casts latents to bf16. We audit releases from three labs. In released checkpoints, fp32 quantization of the shipped latents disagrees with deployed codes on 0.83-1.77% of codes (Falcon-E, BitCPM) and 1.530% (2B-4T); most Falcon-E and BitCPM disagreements are bf16 products exactly on the threshold, which ties-to-even maps to zero, and the unmodified onebitllms exporter reproduces all four Falcon-E releases byte for byte. At fine-tuned endpoints with ULP-matched rates, the documented export lowers greedy GSM8K strict accuracy from 58.79% to 0.78% (Falcon-E-1B-Base) and 36.13% to 0.39% (BitCPM-CANN-0.5B), and a bf16 save and reload lowers 2B-4T's by 27.54 points while its last-number accuracy rises at that endpoint. The strict decrease recurs at each of three new seeds in all three models and has the same direction on 607 reserve test rows, but it is recipe-dependent: at Falcon-E's online-loss-selected rate the export was non-inferior, and after a longer recipe no difference was detected. Two remedies, writing the trained codes directly or adjusting bf16 inputs so unchanged tools emit them, met an inherited 4-point strict-accuracy non-inferiority criterion in all three models, also for the pooled mean over three fixed new trajectories per model and, separately, on the reserve rows. In a frozen study of bf16 latents in fine-tuning (one task, one trajectory per arm), native bf16 writes left 2B-4T's endpoint codes unchanged and changed one Falcon-E code; emulated stochastic-rounding and compensated writers changed codes and were non-inferior to fp32 training at that margin. In two model families, randomized threshold-distance interventions support distance-dependent selection of changed codes.

cs.LG · cs.CL

新论文 · v1arXiv:2610.36269HF 社区推荐

Cross-modal Steering Vectors Exist Across Both Image and Text Generation

Donghao Zeng, Nishant Subramani, Tamar Rott Shaham, Benno Krojer

原文摘要

Are concepts and mechanisms unified across modalities in current AI models, or instead implemented separately for each one? We operationalize this broad question of modality unification by studying cross-modal steering: 1) finding vectors that successfully steer generation on tokens of one modality, which then 2) transfer zero-shot to steer generation when added to tokens of another modality. Our study goes beyond simple correlational alignment (e.g. representational similarity) because steering is fundamentally causal. We study cross-modal steering in several unified multimodal models (UMMs), since they possess shared parameters for language and image generation, focusing on these two modalities. Across diverse steered concepts such as emotion and object size, we show that steering vectors transfer zero-shot in both directions (text-to-image & image-to-text). Even for concepts that are highly unique to one modality (e.g. formal vs. colloquial language), steering vectors often induce a consistent, interpretable effect in the other modality, such as formal clothing and design. This transfer has a geometric signature too: despite a clear modality gap separating text and image representations, steering vectors with strong cross-modal transferability tend to point in similar directions for the same concept. Our findings shed light on how vision and language interact inside unified multimodal models and point towards more general, modality-agnostic approaches for controlling their behavior.

cs.CV · cs.CL · cs.LG

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1868ACL Anthology · ACL 2026

TAIGR: Towards Modeling Influencer Content on Social Media via Structured, Pragmatic Inference

Nishanth Sridhar Nakshatri, Eylon Caplan Purdue University, Rajkumar Pujari Amazon, Dan Goldwasser Purdue University

原文摘要

Health influencers play a growing role in shaping public beliefs, yet their content is often conveyed through conversational narratives and rhetorical strategies rather than explicit factual claims. As a result, claim-centric verification methods struggle to capture the pragmatic meaning of influencer discourse. In this paper, we propose TAIGR (Takeaway Argumentation Inference with Grounded References), a structured framework designed to analyze influencer discourse, which operates in 3 stages: (1) identifying the core influencer recommendation–takeaway; (2) constructing an argumentation graph that captures influencer justification for the takeaway; (3) performing factor graph-based probabilistic inference to validate the takeaway. We evaluate TAIGR on a content validation task over influencer video transcripts on health, showing that accurate validation requires modeling the discourse’s pragmatic and argumentative structure rather than treating transcripts as flat collections of claims.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1947ACL Anthology · ACL 2026

Measuring User’s Mental Models of Speech Translation in Human-AI Collaboration

HyoJung Han Department of Computer Science, University of Maryland, College Park, Nishant Balepur, Jordan Lee Boyd-Graber University of Maryland, College Park, Marine Carpuat University of Maryland, College Park

原文摘要

Millions of people use machine translation (MT) tools daily, yet little is known about their perception of what systems can and cannot do. This paper studies users’ mental models of speech translation systems through a new framework based on cross-lingual question answering, where users either accept MT output or request professional re-translation to answer questions based on the information presented in a foreign language. By analyzing user behavior and accuracy trends across varying translation qualities, we examine to what extent they can predict where the system is likely to be wrong, and how this mental model evolves. Users develop stronger mental models with practice, especially when they have some knowledge of the source language, primarily by relying on surface-level error cues. Moreover, providing speech transcriptions can help users develop better mental models. Our results show the promise of cross-lingual question answering as a downstream task for studying MT mental models and advancing our understanding of human–AI collaboration.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.719ACL Anthology · ACL 2026

BenchMarker: An Education-Inspired Toolkit for Highlighting Flaws in Multiple-Choice Benchmarks

Nishant Balepur, Bhavya Rajasekaran, Hyunjin Jane Oh, Michael Xie, Atrey Desai University of Maryland, College Park and University of Texas at Arlington, Vipul Gupta Scale AI, Steven James Moore George Mason University, Eunsol Choi New York University 等 10 位作者

原文摘要

Multiple-choice question answering (MCQA) is standard in NLP, but benchmarks lack rigorous quality control. We present BenchMarker, an education-inspired toolkit using LLM judges to flag three common MCQ flaws: 1) contamination—items appearing exactly online; 2) shortcuts—cues in the choices that enable guessing; and 3) writing errors—structural/grammatical issues based on a 19-rule education rubric. We validate BenchMarker with human annotations, then run the tool to audit 12 benchmarks, revealing: 2) contaminated MCQs tend to inflate accuracy, while writing errors tend to lower it and change rankings beyond random; and 3) prior benchmark repairs address their targeted issues (i.e., lowering accuracy with LLM-written distractors), but inadvertently add new flaws (i.e. implausible distractors, many correct answers). Overall, flaws in MCQs degrade NLP evaluation, but education research offers a path forward. We release BenchMarker to bridge the fields and improve MCQA benchmark design.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.720ACL Anthology · ACL 2026

Model Internal Sleuthing: Finding Lexical Identity and Inflectional Features in Modern Language Models

Michael Li Carnegie Mellon University, Nishant Subramani Carnegie Mellon University

原文摘要

Large transformer-based language models dominate modern NLP, yet our understanding of how they encode linguistic information relies primarily on studies of early models like BERT and GPT-2. We systematically probe 25 models from BERT Base to Qwen2.5-7B focusing on two linguistic properties: lexical identity and inflectional features across 6 diverse languages. We find a consistent pattern: inflectional features are linearly decodable throughout the model, while lexical identity is prominent early but increasingly weakens with depth. Further analysis of the representation geometry reveals that models with aggressive mid-layer dimensionality compression show reduced steering effectiveness in those layers, despite probe accuracy remaining high. Pretraining analysis shows that inflectional structure stabilizes early while lexical identity representations continue evolving. Taken together, our findings suggest that transformers maintain inflectional features across layers, while trading off lexical identity for compact, predictive representations.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.723ACL Anthology · ACL 2026

Language Models Don’t Know What You Want: Evaluating Personalization in Deep Research Needs Real Users

Nishant Balepur, Malachi Hamada Allen Institute for Artificial Intelligence, Varsha Kishore Allen Institute for Artificial Intelligence, Sergey Feldman Allen Institute for Artificial Intelligence and Data Cowboys, Amanpreet Singh Allen Institute for Artificial Intelligence, Pao Siangliulue Allen Institute for Artificial Intelligence, Joseph Chee Chang Allen Institute for Artificial Intelligence, Eunsol Choi New York University 等 10 位作者

原文摘要

Deep Research (DR) tools (e.g. OpenAI DR) help researchers cope with ballooning publishing counts. Such tools can synthesize scientific papers to answer researchers’ queries, but lack understanding of their users. We change that in MyScholarQA (MySQA), a personalized DR tool that: 1) infers a profile of a user’s research interests; 2) proposes personalized actions for a user’s input query; and 3) writes a multi-section report for the query that follows user-approved actions. We first test MySQA with NLP’s standard protocol: we design a benchmark of synthetic users and LLM judges, where MySQA beats baselines in citation metrics and personalized action-following. However, we suspect this process does not cover all aspects of personalized DR users value, so we interview users in an online version of MySQA to unmask them. We reveal nine nuanced errors of personalized DR undetectable by our LLM judges, and we study qualitative feedback to form lessons for future DR design. In all, we argue for a pillar of personalization that easy-to-use LLM judges can lead NLP to overlook: real progress in personalization is only possible with real users.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-short.23ACL Anthology · ACL 2026

Test-Time Reasoners Are Strategic Multiple-Choice Test-Takers

Nishant Balepur, Atrey Desai University of Maryland, College Park and University of Texas at Arlington, Rachel Rudinger

原文摘要

Large language models (LLMs) now give reasoning before answering, excelling in tasks like multiple-choice question answering (MCQA). Yet, a concern is that LLMs do not solve MCQs as intended, as work finds LLMs sans reasoning succeed in MCQA without using the question, i.e., choices-only. Such partial-input success is often deemed problematic, but reasoning traces could reveal if these strategies are truly shallow in choices-only settings. To study these strategies, reasoning LLMs solve MCQs in full and choices-only inputs; test-time reasoning often boosts accuracy on full and in choices-only half the time. While possibly due to shallow shortcuts, choices-only success is barely affected by the length of reasoning traces, and after finding traces pass faithfulness tests, we show they use less problematic strategies like inferring missing questions. In all, we challenge claims that partial-input success is always a flaw, so we discuss how reasoning traces could separate problematic data from less problematic reasoning.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers) · 2026 · 正式发表

版本更新 · v5arXiv:2403.05530Google Research

Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context

Gemini Team, Petko Georgiev, Ving Ian Lei, Ryan Burnell, Libin Bai, Anmol Gulati, Garrett Tanzer, Damien Vincent 等 1137 位作者

原文摘要

In this report, we introduce the Gemini 1.5 family of models, representing the next generation of highly compute-efficient multimodal models capable of recalling and reasoning over fine-grained information from millions of tokens of context, including multiple long documents and hours of video and audio. The family includes two new models: (1) an updated Gemini 1.5 Pro, which exceeds the February version on the great majority of capabilities and benchmarks; (2) Gemini 1.5 Flash, a more lightweight variant designed for efficiency with minimal regression in quality. Gemini 1.5 models achieve near-perfect recall on long-context retrieval tasks across modalities, improve the state-of-the-art in long-document QA, long-video QA and long-context ASR, and match or surpass Gemini 1.0 Ultra's state-of-the-art performance across a broad set of benchmarks. Studying the limits of Gemini 1.5's long-context ability, we find continued improvement in next-token prediction and near-perfect retrieval (>99%) up to at least 10M tokens, a generational leap over existing models such as Claude 3.0 (200k) and GPT-4 Turbo (128k). Finally, we highlight real-world use cases, such as Gemini 1.5 collaborating with professionals on completing their tasks achieving 26 to 75% time savings across 10 different job categories, as well as surprising new capabilities of large language models at the frontier; when given a grammar manual for Kalamang, a language with fewer than 200 speakers worldwide, the model learns to translate English to Kalamang at a similar level to a person who learned from the same content.

cs.CL · cs.AI