paperlist.ai

AI conference papers

AI conference paper library

Search papers from 30+ top conferences, including NeurIPS, ICLR, ICML, CVPR, ACL and AAAI. Filter by keyword, then open any paper for its abstract, authors and links.

arXiv

A Framework for Low-Effort Training Data Generation for Urban Semantic Segmentation

Damjan Kalšan, Denis Zavadski, Tim Küchler, Haebom Lee, Stefan Roth, Carsten Rother

Synthetic datasets are widely used for training urban scene recognition models, but even highly realistic renderings show a noticeable gap to real imagery. This gap is particularly pronounced when adapting to a specific target domain, such as Cityscapes, where differences in architecture, vegetation, object appearance, and camera characteristics limit downstream performance. Closing this gap with more detailed 3D modelling would require expensive asset and scene design, defeating the purpose of low-cost labelled data. To address this, we present a new framework that adapts an off-the-shelf diffusion model to a target domain using only imperfect pseudo-labels. Once trained, it generates high-fidelity, target-aligned images from semantic maps of any synthetic dataset, including low-effort sources created in hours rather than months. The method filters suboptimal generations, rectifies image-label misalignments, and standardises semantics across datasets, transforming weak synthetic data into competitive real-domain training sets. Experiments on five synthetic datasets and two real target datasets show segmentation gains of up to +8.0%pt. mIoU over state-of-the-art translation methods, making rapidly constructed synthetic datasets as effective as high-effort, time-intensive synthetic datasets requiring extensive manual design. This work highlights a valuable collaborative paradigm where fast semantic prototyping, combined with generative models, enables scalable, high-quality training data creation for urban scene understanding.

arXiv

Egosurg: Arbitrary view synthesis for egocentric replay of operating room workflows from ambient cameras

Han Zhang, Lalithkumar Seenivasan, Jose L. Porras, Roger D. Soberanis-Mukul, Hao Ding, Hongchao Shu, Benjamin D. Killeen, Ankita Ghosh, Lonny Yarmus, Jeffrey K. Jopling, Masaru Ishii, Angela C. Argento, Mathias Unberath

Observing surgical practice has historically relied on fixed vantage points or recollections, leaving the egocentric perspectives that shape clinical decisions undocumented. Ambient fixed cameras capture the operating room (OR) at room scale but cannot recover what any individual team member actually saw. We present EgoSurg, a framework that reconstructs dynamic OR scenes from sparse wall-mounted stereo video and renders arbitrary, role-specific egocentric views without instrumenting personnel or interfering with clinical workflow. EgoSurg initializes a per-timestamp 3D Gaussian Splatting representation from scale-aware stereo depth and refines it with an image-conditioned diffusion model that corrects auxiliary rendered views, mitigating artifacts caused by limited camera coverage, crowding, and occlusion. We evaluated the framework on four real robotic pulmonology procedures and two simulated full-workflow sessions across two hospital sites. Near-field reconstruction fidelity was consistent (PSNR 26.8~dB, SSIM .895) across five workflow phases and both sites, and synthesized egocentric views reached a PSNR of 17.8~dB and an SSIM of .766 against paired hand-held point-of-view recordings. We further demonstrate three case studies for intended use: adjudicating a simulated sterile field violation, replaying a procedure from role-specific viewpoints, and testing a counterfactual change in personnel position. These results indicate that existing ambient camera infrastructure can be turned into a navigable 3D record of surgical work, supporting retrospective review of safety events, training, and workflow analysis from every angle.

arXiv

Gaze Authentication: Factors Influencing Authentication Performance

Dillon Lohr, Michael J Proulx, Mehedi Hasan Raju, Oleg V. Komogortsev

This paper examines the key factors that influence the performance of state-of-the-art gaze-based authentication. Experiments were conducted on a large-scale, in-house dataset comprising 8,849 subjects collected with Meta Quest Pro equivalent hardware running a video oculography-driven gaze estimation pipeline at 72~Hz. State of the neural network architecture was employed to study the influence of the following factors on authentication performance: eye tracking signal quality, various aspects of eye tracking calibration, and simple filtering on estimated raw gaze. This report provides performance results and their analysis.

arXiv

High-Frequency First: A Two-Stage Approach for Improving Image INR

Sumit Kumar Dam, Mrityunjoy Gain, Eui-Nam Huh, Choong Seon Hong

Implicit Neural Representations (INRs) have emerged as a powerful alternative to traditional pixel-based formats by modeling images as continuous functions over spatial coordinates. A key challenge, however, lies in the spectral bias of neural networks, which tend to favor low-frequency components while struggling to capture high-frequency (HF) details such as sharp edges and fine textures. While prior approaches have addressed this limitation through architectural modifications or specialized activation functions, the potential of complementary training strategies remains relatively underexplored. In this work, we propose an orthogonal direction by directly guiding the training process. Specifically, we introduce a two-stage training strategy where a neighbor-aware soft mask adaptively assigns higher weights to pixels with strong local variations, encouraging early focus on fine details. The model then transitions to full-image training. Experimental results show that our approach is complementary to existing INR methods and consistently improves reconstruction quality. By assigning frequency-aware importance to pixels in image INR, our work offers an effective avenue to mitigate the spectral bias problem.

arXiv

DOD-SA: Infrared-Visible Decoupled Object Detection with Single-Modality Annotations

Hang Jin, CHENQIANG GAO, Junjie Guo, Fangcen Liu, Qinyao Chang, Kanghui Tian, Deyu Meng

Infrared-visible object detection has shown great potential in real-world applications, enabling robust all-day perception by leveraging the complementary information of infrared and visible images. However, existing methods typically require dual-modality annotations to output decoupled detection results, leading to high annotation costs and limiting scalability in large-scale remote sensing applications. To address this challenge, we propose a novel infrared-visible \textbf{D}ecoupled \textbf{O}bject \textbf{D}etection framework with \textbf{S}ingle-modality \textbf{A}nnotations, called DOD-SA. It is built upon a Single- and Dual-Modality Collaborative Teacher-Student Network (CoSD-TSNet), which consists of a single-modality branch (SM-Branch) and a dual-modality decoupled branch (DMD-Branch). This design enables cross-modality knowledge transfer from the labeled modality to the unlabeled modality, and facilitates effective cross-branch supervision. To further improve the quality of pseudo-labels, we introduce a Progressive and Self-Tuning Training Strategy (PaST) that trains the model in three stages: 1) SM-Branch self-training, 2) SM-Branch guiding the learning of DMD-Branch, and 3) DMD-Branch refinement. In addition, we design a Pseudo Label Assigner (PLA) to match labels across modalities, explicitly addressing modality misalignment during training.

arXiv

MObyGaze: a film dataset of multimodal objectification densely annotated by experts

Elisa Ancarani, Julie Tores, Lucile Sassatelli, Hui-Yin Wu, Remy Sun, Paul Mouret, Clement Bergman, Lea Andolfi, Victor Ecrement, Thierry Devars, Magali Guaresi, Virginie Julliard, Sarah Lecossais, Frederic Precioso

Characterizing and quantifying gender representation disparities in audiovisual storytelling contents is necessary to grasp how stereotypes may perpetuate on screen. In this article, we consider the high-level construct of objectification and introduce a new AI task to the ML community: characterize and quantify complex multimodal (visual, speech, audio) temporal patterns producing objectification in films. Building on film studies and psychology, we define the construct of objectification in a structured thesaurus involving 5 sub-constructs manifesting through 11 concepts spanning 3 modalities. We introduce the Multimodal Objectifying Gaze (MObyGaze) dataset, made of 20 movies annotated densely by experts for objectification levels and concepts over freely delimited segments: it amounts to 6072 segments over 43 hours of video with fine-grained localization and categorization. We formulate new video interpretation tasks, show the feasibility of multimodal objectification detection, and analyze data and model bias to propose improvements. We exemplify two applications of MObyGaze, showing how rich concept annotation can improve model reliability and explainability. We make our code and our dataset available to the community and described in the Croissant format: https://github.com/husky-helen/MObyGaze.

arXiv

PiT: Progressive Diffusion Transformer

Jiafu Wu, Yabiao Wang, Jian Li, Jinlong Peng, Yun Cao, Chengjie Wang, Jiangning Zhang

Diffusion Transformers (DiTs) achieve remarkable performance within image generation via the transformer architecture. Conventionally, DiTs are constructed by stacking serial isotropic global modeling transformers, which face significant quadratic computational cost. However, through empirical analysis, we find that DiTs do not rely as heavily on global information as previously believed. In fact, most layers exhibit significant redundancy in global computation. Additionally, conventional attention mechanisms suffer from low-frequency inertia, limiting their efficiency. To address these issues, we propose Pseudo Shifted Window Attention (PSWA), which fundamentally mitigates global attention redundancy. PSWA achieves moderate global-local information through window attention. It further utilizes a high-frequency bridging branch to simulate shifted window operations, which both enrich the high-frequency information and strengthen inter-window connections. Furthermore, we propose the Progressive Coverage Channel Allocation (PCCA) strategy that captures high-order attention without additional computational cost. Based on these innovations, we propose a series of Pseudo Progressive Diffusion Transformer (PiT). Our extensive experiments show their superior performance; for example, our proposed PiT-L achieves 54% FID improvement over DiT-XL/2 while using less computation.

arXiv

Temporally-Grounded Language Generation: Towards Real-Time Vision-Language Models

Keunwoo Peter Yu, Joyce Chai

Vision-language models (VLMs) have shown remarkable progress in offline tasks such as image captioning and video question answering. However, real-time interactive environments impose new demands on VLMs, requiring them to generate utterances that are not only semantically accurate but also temporally precise. We identify two core capabilities necessary for such settings---\textit{perceptual updating} and \textit{contingency awareness}---and propose a new benchmark task, \textbf{Temporally-Grounded Language Generation (TGLG)}, to evaluate them. TGLG requires models to generate utterances in response to streaming video such that both content and timing align with dynamic visual input. To support this benchmark, we curate evaluation datasets from sports broadcasting and egocentric human interaction domains, and introduce a new metric, \textbf{TRACE}, to evaluate TGLG by jointly measuring semantic similarity and temporal alignment. Finally, we present \textbf{Vision-Language Model with Time-Synchronized Interleaving (VLM-TSI)}, a model that interleaves visual and linguistic tokens in a time-synchronized manner, enabling real-time language generation without relying on turn-based assumptions. Experimental results show that VLM-TSI significantly outperforms a strong baseline, yet overall performance remains modest---highlighting the difficulty of TGLG and motivating further research in real-time VLMs. Our code is available at https://github.com/yukw777/tglg.

arXiv

Federated Adversarial Training with Transformers

Ahmed Aldahdooh, Wassim Hamidouche, Olivier Déforges

Federated learning (FL) has emerged to enable global model training over distributed clients' data while preserving its privacy. However, the global trained model is vulnerable to the evasion attacks especially, the adversarial examples (AEs), carefully crafted samples to yield false classification. Adversarial training (AT) is found to be the most promising approach against evasion attacks and it is widely studied for convolutional neural network (CNN). Recently, vision transformers have been found to be effective in many computer vision tasks. To the best of the authors' knowledge, there is no work that studied the feasibility of AT in a FL process for vision transformers. This paper investigates such feasibility with different federated model aggregation methods and different vision transformer models with different tokenization and classification head techniques. In order to improve the robust accuracy of the models with the not independent and identically distributed (Non-IID), we propose an extension to FedAvg aggregation method, called FedWAvg. By measuring the similarities between the last layer of the global model and the last layer of the client updates, FedWAvg calculates the weights to aggregate the local models updates. The experiments show that FedWAvg improves the robust accuracy when compared with other state-of-the-art aggregation methods.

arXiv

ReVA: A Region-Aware Visual Assistant for Visually Grounded Question Answering

Anoop Senthil

Multimodal Large Language Models (MLLMs) have achieved remarkable progress in Visual Question Answering (VQA), yet they continue to struggle with questions requiring precise spatial reasoning and fine-grained visual understanding. These limitations often manifest as object, attribute, and spatial hallucinations, where models generate confident but visually unsupported responses due to insufficient region-level and fine-grained visual grounding. To address this challenge, we propose ReVA, a region-aware VQA model that employs a frozen CLIP ViT-L/14 Vision Transformer (ViT) and a Qwen2.5-7B-Instruct large language model (LLM) connected through a dual bridge that aligns both whole-image and region-level representations with the LLM's embedding space. The image bridge maps final transformer block features into image tokens. The region bridge maps cropped features from enriched intermediate features across ViT blocks so early texture and later object cues are more evident, into K region tokens for every bounding box. ReVA uses a detector stack that supplies automatic zero-shot bounding boxes that are both question-agnostic and question-dependent, using RAM++ (Recognize Anything Model), spaCy, and Grounding DINO. The image tokens and region tokens are concatenated as an LLM prompt prefix to jointly encode scene-level context and fine-grained regional evidence when answering questions. Evaluated on VQAv2, MMBench, POPE, and SEED-Bench, ReVA achieves 82.85% mean F1 on POPE, compared with 81.14% for an image-token baseline without region tokens. These results demonstrate that explicit region-aware visual representations reduce object hallucination and improve the factual grounding of MLLMs.

arXiv

Compositional Failure in Audio-Visual LLMs: Late-Layer Prior Dominance Under Cross-modal Conflict

Adarsh Sudheer, David Li, Omar Elbanna, Ishaan Kodarapu, Arjun Bahuguna, Vasu Sharma

We study audio-visual conflict as a compositional generalization test for AV-LLMs: the model must combine synchronized but semantically incompatible audio and video evidence and decide whether the pair matches. On VideoLLaMA 2-7B-AV, three alignment configurations remain nearchance on the scored exact-string Yes/No subset of AVHBench, even though their output priors shift substantially. Similarly, off-the-shelf InternVideo2 experienced a 32.3% accuracy decrease specifically under cross-modal conflict, accompanied by a 17.3% instruction-following failure. We call this failure mode prior dominance: late-layer commitment to an internally preferred answer pattern that is weakly grounded in the conflicting inputs. To explain this behavior, we conduct a mechanistic interpretability analysis and find that commitment remains concentrated at 25.5 $\pm$ 1 layers. We show that stronger temporal alignment changes answer bias, but do not improve compositional conflict resolution. Code and data to reproduce our mechanistic audit and behavioral evaluations are available at https://github.com/AdarshSudheer09/AVHBench-dmai.

arXiv

SURE-Challenge: Evaluating Speech Evidence Before Speech-LLM Generation

Mengzhe Geng

Speech LLMs are usually graded after they answer, although an operating system first has to decide whether to send a waveform to the model. We define the Speech-Unsupported Rejection Evaluation Challenge (SURE-Challenge) for this admission step. The benchmark pairs LibriSpeech-derived transcription and first-word question answering with unsupported silence, colored noise, synthetic tones, and source-ambiguous babble under disjoint source splits. Front-end ablations use Qwen2-Audio; the selected energy-plus-Whisper-score rule is then replayed before six speech/audio LLMs. On the leakage-screened 474-example SURE-Extended test set, raw Qwen2-Audio rejects 15/204 unsupported inputs, whereas the fixed rule rejects 196/204 and leaves supported accuracy unchanged. External evaluations qualify this result: Common Voice retention drops as the Whisper-score threshold is tightened, and no-speed babble gives 18 to 24 rejected clips out of 54 across regenerated seeds. The result identifies a pre-generation error mode missed by answer-only scoring.

arXiv

Memorization Is Not Extraction: Tight Differential-Privacy Bounds and Audit Blind Spots

Xujun Che, Depeng Xu, Shuhan Yuan

Memorization in large language models is measured through a zoo of definitions whose formal relations are unknown, and differential privacy (DP) is treated as a proxy against all of them at once. We pin down the exact DP constant for the two that carry the practical weight, counterfactual memorization and adaptive extraction, and show that they do not control each other. Under $f$-DP, every adaptive extraction protocol with list budget $m$ succeeds with probability at most $1-f(κ)$ for the oblivious baseline $κ$, and the bound is tight on a dense set of baselines: DP uniformly controls extraction exactly up to a threshold in how well the secret can be guessed a priori. Min-entropy certifies that baseline distribution-free, since $H_\infty\geε\log_2 e+\log_2(m/τ)$ holds extraction below a risk level $τ\le1/2$ under pure $ε$-DP for every prior, and is exact on uniform priors. On the memorization side, $f$-DP caps the counterfactual memorization of any bounded score at an advantage functional $η(f)$, equal to $\tanh(ε/2)$ under pure DP; for $k\ge2$ duplicated copies the naive $ε\mapsto kε$ bound $\tanh(kε/2)$ is unattainable, the exact constant being a closed-form staircase attained by geometric noisy counting. That cap is attained inside the local score class used in practice, and it is there that the two measures separate: one mechanism is memorized yet unextractable, another fully extractable yet exactly invisible to every loss-based score. The two-sided blind spot this opens for loss-based auditing and unlearning verification survives on billion-parameter models: a reserved-trigger release is recovered verbatim from one prompt while the audits practitioners deploy certify it clean.

arXiv

Why Didn't It Check? Unsupported Final Claims and Their Repair in Two Tool-Equipped Language Models

Justin Bronder

A language model with access to tools can commit to a final claim unsupported by the evidence it has seen, even when a single available tool call would resolve the uncertainty and its instructions explicitly forbid assumptions and guesses. We separate this failure into two precisely defined quantities: occurrence, how often the model makes an unsupported claim on its own, measured from the visible evidence and final claim without using the hidden correct answer; and conditional repair, how often those same naturally occurring unsupported claims are repaired when the missing evidence is supplied. On one fixed Qwen3-32B setup, 33 of 512 first responses to 256 new prompt templates ended with an unsupported established claim. We replayed each case from an exact copy of the state in which the claim occurred; within each matched replay, the alternative tool responses had the same structure and length and differed only in a one-character response code. Resolving evidence repaired 33 of 33 claims; a matched response carrying no useful information repaired 0 of 33. When the evidence supported the original answer, the model preserved 33 of 33, with no observed harm. In a separate experiment, on 64 cases where evidence was needed, an automatic checking rule added 21 evidence calls, corrected all 10 wrong unsupported claims, preserved the 11 that were correct by accident, and never changed a correct answer into a wrong one. On a fixed Gemma 4 setup using the same sampling settings, the model called the tool in all 512 first responses and never made an unsupported final claim, so conditional repair could not be measured for that setup. These results describe two local fixed model setups on two synthetic task families. They do not show how common this failure is in real-world deployments, nor that it reflects a general mechanism shared across models.

arXiv

Fast Weight Attention for Continual Learning

Yifan Zhang, Steve Ta, Jasper Zhang, Jichen Feng, Shuzhen Li, Yongxin Zhang, Yifeng Liu, Huizhuo Yuan, Mengdi Wang, Quanquan Gu, Andrew Chi-Chih Yao

Recurrent fast-weight memories and selective state-space models compress an expanding context into a fixed-size recurrent state, making the state transition an online learning rule. We study this rule under read-after-write autoregressive semantics. For the prefix-prediction objective considered here, the local fast-memory example revealed at step $t$ is the prefix-aligned pair $(\mathbf{x}_t,\mathbf{y}_t)=(ϕ(\mathbf{k}_{t-1}),\mathbf{v}_t)$. The common same-step association $(ϕ(\mathbf{k}_t),\mathbf{v}_t)$ remains causal, but optimizes a different internal objective. We derive normalized first-order updates for squared-error regression and negative inner-product objectives. The regression family comprises Falcon-1 (a scalar NLMS update), Falcon-2 (its per-column extension), and Falcon-3 (a sliding-window mini-batch update); Falcon-1A/Falcon-2A/Falcon-3A are the corresponding inner-product variants. We provide recurrent, masked-parallel, and chunk-parallel forms, together with numerically stable positive-decay renormalization. Representative variants remain competitive in language modeling and improve length extrapolation on variable-digit addition. This framework separates temporal alignment, plasticity, forgetting, and bounded rehearsal in recurrent sequence models.

arXiv

Informational Antilocality and the Locality Bias in LLMs

Andrew McInnerney, Shane Storks, Steven Abney, Richard L Lewis

We consider the ability of transformer-based language models (LLMs) to learn what we call k-antilocal languages, i.e., languages that have no mutual information across any span of $k$ contiguous symbols. We construct such languages with increasing $k$, finding that LLMs trained on them achieve comparable cross-entropy loss regardless of antilocality, but converge more slowly on more antilocal languages. Our findings support the idea that non-local dependencies are more difficult to learn, but the evidence for this bias comes from learning speed rather than learning success.

arXiv

Load-Bearing Context: The Question Damage Score for Evaluating Context Reliance in Linguistic Reasoning

Neh Majmudar, Elena Filatova

Determining whether large language models derive answers from context or prior knowledge remains a fundamental challenge. Self-contained linguistic olympiad puzzles provide a controlled setting where all answers derive solely from expert-designed context examples without external knowledge. Removing individual context examples can eliminate information needed for specific questions while leaving the rest of the puzzle unchanged. We leverage this to introduce a diagnostic framework for analyzing individual context examples. Using 53 UK Linguistics Olympiad puzzles, we generate two modified variants by deleting a single context example: (1) uniform random deletion, and (2) targeted deletion (inspired by error-correcting codes) to remove a structurally load-bearing example uniquely carrying necessary information. We formalize this impact using a Question Damage Score to classify puzzles as fragile or robust. Evaluating three frontier LLMs under instructions to abstain when information is insufficient, we find they rarely abstain, often continuing to produce correct answers after load-bearing context is removed. These findings motivate further investigation into context-based reasoning, prior knowledge, memorization, and linguistic inference. Beyond abstention, the framework enables fine-grained analyses of context reliance, including causal interventions, stopping-set analysis, targeted contamination studies, and mechanistic interpretability.

arXiv

The Calls are Coming from Inside the Model: Investigating Probe-based Detection of Tool-Calling Errors in LLMs

Eric Yeats, Brendan Kennedy, Loc Truong, John Buckheit, Jung Lee, Jesse Friedbaum, John Emanuello, Henry Kvinge

The hidden states of large language models (LLMs) are known to capture rich information relating to model knowledge and behavior that can be hard to extract from examination of input and output alone. As LLM-based systems increasingly interface with the external world, one area of concern is detecting incorrect or improper use of tools. Motivated by this, we study the effectiveness of using linear probes to detect incorrect tool-calls, measuring probe efficacy across 18 tool-calling LLMs evaluated on the Berkeley Function Calling Leaderboard. Overall, we find that probing is an effective means to catch a range of different tool-calling errors, including errors arising from using an argument that has the wrong value but the correct type, which might not be recorded by standard logging frameworks. Important factors in success include model size, probing layer, and model post-training type. We also show that probes are capable of generalizing to novel types of errors, which is critical in real world deployments.

arXiv

Below the Noise Floor: Bimodal Seed Collapse and Distinct Failure Modes in Small-Model Knowledge Distillation

Dipto Sumit, Sakib Ul Haque, Farig Sadeque

Function routing -- selecting the correct API call from a fixed catalog given a natural-language request -- is a deployment problem where small students are attractive but knowledge distillation gains are typically reported single-seed, at scales where seed variance is unknown. On a 740-instance healthcare API routing task with a 1.5B Qwen student and a 20B teacher, we compare eight KD variants against supervised cross-entropy, using three to six seeds for key configurations. We find: (i) per-seed standard deviation ranges from 2.8 to 48.7 percentage points, swallowing every claimed KD gain below five points; (ii) three of seven KD variants exhibit bimodal collapse, with at least one in three to five seeds falling below 55% accuracy while the others train normally, and a fourth showing elevated variance; (iii) collapse has distinct modes -- wrong-function selection for ce_kd and ce_paraphrase, and a previously undocumented output-truncation mode for reasoning_kd, where the model emits reasoning but terminates before producing a function name (0.9% accuracy); (iv) only progressive_kd and rank_kd avoid collapse across observed seeds, with sigma <= 3.9 pp; (v) a naive cross-split +3.78 pp gain from input enrichment reverses to -2.70 pp under controlled within-split multi-seed re-testing. Single-seed evaluation is therefore unable to detect central failure modes in small-model KD.

arXiv

First Make It Playable, Then Make It Good: Staged Interaction Learning for Small Dialogue-Game Agents

Syed Mahbubul Huq, Pranava Madhyastha

We present Qwen-GuidePlay-2B, a 2B-parameter language model for dialogue-game interaction. We fine-tune Qwen3.5-2B using three steps: a) SFT on only successful game trajectories from Playpen, b) weighted turn-level SFT, and c) teacher-guided SFT. The teacher model (which is a larger model) is only used to fix formatting and evaluate examples, but does not create new gold actions. Our final model scores 57.12 clemscore and 42.68 statscore on the public Playpen validation. In the officially released challenge results, our model obtains the second-highest Playpen clemscore delta among submitted systems (which is approximately +36 over its base model). Our findings suggest that imitating full trajectories helps with playability, while turn-level and teacher-guided training usually improve decision-making and increase the overall score. Alternative procedurally heavy approaches like replay-repair and hard-example mining did not help, which suggests that small models are performant simply by using careful curation strategies rather than aggressive changes. We make available both the model and the code for reproducibility.

Keep browsing

More papers from top AI conferences

30+ top conferences, by conference, year, and topic.