arxiv_id large_stringlengths 10 10 | title large_stringlengths 14 192 | abstract large_stringlengths 73 1.92k | published_at large_stringdate 2026-05-21 00:00:00 2026-05-29 00:00:00 | categories listlengths 1 6 | abstract_embedding listlengths 384 384 |
|---|---|---|---|---|---|
2605.31604 | Representation Forcing for Bottleneck-Free Unified Multimodal Models | Unified multimodal models (UMMs) aim to handle perception and generation in a single model. Yet existing UMMs still rely on a frozen, separately pretrained VAE for image generation, imposing a structural bottleneck. Naively removing it introduces a quality gap, as the model must learn both high-level structure and low-... | 2026-05-29 | [
"cs.CV"
] | [
-0.04890205338597298,
-0.10049933940172195,
0.06791713088750839,
0.06713556498289108,
0.07446564733982086,
-0.020450303331017494,
-0.06765308231115341,
-0.04907295107841492,
0.0010043798247352242,
-0.05567323416471481,
-0.06212345138192177,
-0.12359961867332458,
0.008114486001431942,
0.026... |
2605.31603 | Lumos-Nexus: Efficient Frequency Bridging with Homogeneous Latent Space for Video Unified Models | Connector-based video unified models have demonstrated strong capability in instruction-grounded video synthesis, but integrating a large high-fidelity generator into the unified training loop is computationally prohibitive, limiting achievable visual quality. We therefore propose Lumos-Nexus, a training-efficient unif... | 2026-05-29 | [
"cs.CV",
"cs.AI"
] | [
0.01349659264087677,
-0.12581884860992432,
-0.011904564686119556,
0.0234975628554821,
0.06542549282312393,
-0.03824600204825401,
-0.06469496339559555,
-0.04221169278025627,
0.0350177139043808,
-0.04756224900484085,
-0.06504309177398682,
-0.0755668506026268,
-0.014029874466359615,
0.0417586... |
2605.31598 | Linear Scaling Video VLMs for Long Video Understanding | Video vision-language models (VLMs) are increasingly used in long-horizon and streaming settings, yet most video encoders still rely on spatiotemporal self-attention, causing compute and latency to grow quadratically with the number of frames. Existing efficiency methods improve scalability but often lose accuracy rela... | 2026-05-29 | [
"cs.CV"
] | [
0.030256133526563644,
-0.10650993138551712,
-0.0023578880354762077,
-0.03717275708913803,
0.09389487653970718,
0.07130727916955948,
-0.009465117938816547,
-0.02875298634171486,
0.051415301859378815,
-0.06286777555942535,
-0.07372935861349106,
-0.041976071894168854,
-0.08383338153362274,
0.... |
2605.31597 | SOCO: Benchmarking Semantic Object Correspondence in Vision Foundation Models | Measuring structured object understanding in vision foundation models remains challenging due to inconsistent evaluation protocols and limited part-level supervision. Semantic correspondence (SC) evaluates this capability by testing whether object parts can be matched across instances and categories under large variati... | 2026-05-29 | [
"cs.CV"
] | [
0.05323132500052452,
-0.05262693762779236,
0.023284094408154488,
-0.010524310171604156,
0.2128268927335739,
-0.045096661895513535,
-0.029002360999584198,
-0.021857554093003273,
0.02888568677008152,
-0.06735334545373917,
-0.06957440078258514,
-0.09016406536102295,
0.0218961238861084,
0.0728... |
2605.31596 | KLIP: localized distribution shift detection via KL-divergence with diffusion priors in Inverse Problems | Diffusion models have shown promising performance as data-driven priors for computational imaging, as well as some capacity to detect out-of-distribution (OOD) images. However, existing approaches to OOD detection often require some knowledge of the shifted distribution, fail to detect subtle or localized distribution ... | 2026-05-29 | [
"cs.CV",
"cs.LG"
] | [
0.0007767697097733617,
-0.1185343861579895,
0.06368280202150345,
-0.028583141043782234,
0.07981376349925995,
-0.14199504256248474,
0.03090251237154007,
-0.07962967455387115,
0.09785347431898117,
-0.011132418178021908,
0.052862487733364105,
-0.008812354877591133,
0.0006479594740085304,
0.06... |
2605.31595 | Learning Global Motion with Compact Gaussians for Feed-Forward 4D Reconstruction | Dynamic scene reconstruction from monocular video remains a fundamental challenge in computer vision. Existing feed-forward methods predict 3D Gaussians pixel-wise for each frame, suffering from duplicated Gaussians and view-dependent biases that hinder effective learning of scene motion. We present C4G, a feed-forward... | 2026-05-29 | [
"cs.CV"
] | [
-0.008206839673221111,
-0.08620259165763855,
0.0758843645453453,
0.020194148644804955,
0.0715818777680397,
-0.01659989356994629,
-0.028923703357577324,
-0.07126185297966003,
-0.003907033242285252,
-0.02479085698723793,
-0.06063622981309891,
0.0016998484497889876,
-0.0305052287876606,
0.067... |
2605.31594 | A Tight Theory of Error Feedback Algorithms in Distributed Optimization | Communication costs are a major bottleneck in distributed learning and first-order optimization. A common approach to alleviate this issue is to compress the gradient information exchanged between agents. However, such compression typically degrades the convergence guarantees of gradient-based methods. Error feedback m... | 2026-05-29 | [
"cs.LG",
"math.OC"
] | [
0.015303336083889008,
-0.007927833124995232,
-0.011652971617877483,
-0.005961260292679071,
0.028871458023786545,
-0.02186768874526024,
-0.06758943200111389,
-0.0029188133776187897,
0.03030405379831791,
0.026563262566924095,
-0.025951558724045753,
0.08747871220111847,
0.053081970661878586,
... |
2605.31593 | Stateful Online Monitoring Catches Distributed Agent Attacks | Language models can find thousands of severe software vulnerabilities, and agents are increasingly being misused for cyberattacks. To avoid detection, attackers frequently distribute their misuse, splitting a harmful task across many user accounts so each individual transcript looks benign. Because safety monitors scor... | 2026-05-29 | [
"cs.CR",
"cs.AI"
] | [
-0.0033580136951059103,
-0.06643178313970566,
-0.10122478008270264,
-0.010895520448684692,
0.07383669167757034,
0.03975769877433777,
0.06856062263250351,
-0.04421459883451462,
0.05698155239224434,
-0.03269936144351959,
-0.035863980650901794,
-0.0955902487039566,
0.11781803518533707,
0.0345... |
2605.31591 | CoFiDA-M: Concept-Aware Feature Modulation for Cross-Domain Adaptation with Image-Only Inference | Models for AI-based skin cancer screening suffer a severe performance drop when shifting from expert dermoscopic (source) images to consumer-grade clinical (target) images, hindering real-world deployment. Existing domain adaptation methods often ignore crucial semantic invariants, such as clinical concepts. While new ... | 2026-05-29 | [
"cs.CV"
] | [
0.0213592778891325,
-0.028271855786442757,
-0.011042304337024689,
0.005663301795721054,
0.13948537409305573,
-0.051638826727867126,
0.05487581714987755,
-0.0008911151089705527,
-0.02978554554283619,
-0.02032320201396942,
0.009448162280023098,
-0.08479920774698257,
0.04367006942629814,
0.09... |
2605.31590 | TunerDiT: Training-free Progressive Steering of Diffusion Transformer for Multi-Event Video Generation | Text-to-video (T2V) generation faces challenging questions when generating videos with long horizons containing multiple events. Inspired by the intrinsics of the diffusion process, we probe video diffusion transformers (DiTs) and uncover intrinsic turning points in the DiT denoising trajectory where conditioning text ... | 2026-05-29 | [
"cs.CV",
"cs.AI"
] | [
0.011719529516994953,
-0.07370950281620026,
0.05269576609134674,
0.014665662311017513,
0.12212522327899933,
-0.011980608105659485,
-0.05578267201781273,
-0.05913974344730377,
0.09088017791509628,
-0.0801362544298172,
-0.09276596456766129,
-0.055061519145965576,
-0.0787789598107338,
0.03446... |
2605.31589 | Recognizing Co-Speech Gestures in-the-Wild | While humans naturally gesture during speech, only a sparse subset of these movements are visually depictive and semantically linked to specific spoken words. Current multimodal models struggle to capture these semantic co-speech gestures, heavily bottlenecked by a lack of precisely annotated training data. To address ... | 2026-05-29 | [
"cs.CV"
] | [
0.06726762652397156,
-0.1788729429244995,
0.04424851015210152,
-0.06433066725730896,
0.138288676738739,
-0.0016880226321518421,
-0.006116632372140884,
-0.07591208815574646,
-0.018806342035531998,
-0.11130432039499283,
-0.03435979038476944,
-0.07189197093248367,
-0.0661928653717041,
0.10114... |
2605.31586 | Language Models Learn Constructional Semantics, Not To Mention Syntax: Investigating LM Understanding of Paired-Focus Constructions | Grasping the semantics of rare constructions (form-meaning pairings) has been shown to be a challenging problem that has currently only been solved by the largest LLMs. It remains an open question if open-source models have robust constructional understanding, and if so, what learning dynamics underlie the acquisition ... | 2026-05-29 | [
"cs.CL",
"cs.AI"
] | [
0.03948335722088814,
-0.10472419857978821,
0.02872558683156967,
0.08640040457248688,
0.057043202221393585,
-0.010057641193270683,
0.014720838516950607,
0.023683913052082062,
0.06041277199983597,
-0.040891993790864944,
-0.005981145892292261,
-0.046318791806697845,
0.058956895023584366,
-0.0... |
2605.31584 | LongTraceRL: Learning Long-Context Reasoning from Search Agent Trajectories with Rubric Rewards | Long-context reasoning remains a central challenge for large language models, which often fail to locate and integrate key information in extensive distracting content. Reinforcement learning with verifiable rewards (RLVR) has shown promise for this task, yet existing methods are limited by low-confusability distractor... | 2026-05-29 | [
"cs.CL",
"cs.AI",
"cs.LG"
] | [
-0.07732322067022324,
-0.03168017044663429,
-0.0009500112500973046,
0.06244305893778801,
0.07751091569662094,
-0.003169669769704342,
0.053800247609615326,
-0.012801295146346092,
0.03489002585411072,
0.029950061812996864,
-0.015670685097575188,
-0.07781223207712173,
0.05453195050358772,
0.0... |
2605.31581 | Choosing the Lens: Strategic Perspective Activation in Context-Dependent Argumentation | The same arguments often need to be evaluated under different external regimes. An agent with influence over the regime has a strategic lever that standard formalisms do not directly capture. We introduce context-dependent argumentation frameworks (CDAFs), an extension of Dung's theory in which a defeat function determ... | 2026-05-29 | [
"cs.AI"
] | [
0.02975156344473362,
0.0030282684601843357,
-0.014780810102820396,
-0.0620485357940197,
0.045198068022727966,
0.056673090904951096,
0.10556472092866898,
0.06608125567436218,
0.08672742545604706,
0.021660158410668373,
-0.07744164019823074,
-0.043570663779973984,
0.046837784349918365,
0.0118... |
2605.31580 | Giving Sensors a Voice: Multimodal JEPA for Semantic Time-Series Embeddings | Transformer-based architectures have advanced sequence modeling in language and vision, yet general-purpose representation learning for heterogeneous multivariate time series remains underexplored. We introduce CHARM (Channel-Aware Representation Model), which incorporates channel-level textual descriptions into a Tran... | 2026-05-29 | [
"cs.LG"
] | [
-0.07730375975370407,
-0.1484818458557129,
0.08703596144914627,
0.024224750697612762,
0.036517463624477386,
0.05599038302898407,
-0.02337258867919445,
-0.03239136189222336,
0.06330569088459015,
-0.10171207040548325,
-0.08317982405424118,
-0.0208351518958807,
-0.016047410666942596,
0.091607... |
2605.31577 | SurGe: Improved Surface Geometry in Point Maps | Recent feedforward 3D reconstruction methods predict point maps and estimate global 3D geometry remarkably well. However, their predictions still exhibit inaccurate local surface geometry, which is clearly visible qualitatively but only weakly reflected in common metrics. To make these errors more explicit in evaluatio... | 2026-05-29 | [
"cs.CV"
] | [
-0.021745705977082253,
-0.08499537408351898,
0.08285635709762573,
-0.015239046886563301,
0.0890154168009758,
-0.01975647546350956,
-0.06385335326194763,
-0.061358630657196045,
-0.014095814898610115,
0.0018488147761672735,
-0.02623005025088787,
-0.003682701848447323,
0.031217342242598534,
0... |
2605.31576 | Joint Multi-Camera LiDAR Extrinsic Calibration via Learned Pairwise Initialization and Geometric Refinement | Most learning-based camera-LiDAR calibration methods treat each camera-LiDAR pair independently, ignoring the rigid geometric coupling in multi-camera platforms. As a result, per-camera estimates may be individually accurate yet inconsistent at the system level. We present a two-stage framework for joint multi-camera L... | 2026-05-29 | [
"cs.CV"
] | [
-0.006601563189178705,
-0.05542684346437454,
-0.04628445580601692,
-0.04019153490662575,
0.09446609765291214,
-0.026124682277441025,
0.026805967092514038,
-0.04152790457010269,
0.019638344645500183,
-0.01040622964501381,
-0.02787180058658123,
-0.06575334817171097,
-0.009164025075733662,
0.... |
2605.31575 | SPECTRA: Synthetic IR Test Collections with Relevance Oracles and Controlled Distractor Diagnostics | Scalable information retrieval testing needs corpora that are large enough to stress index construction, ranking latency, query routing, and evaluation tooling, yet human-judged test collections remain expensive and may be unavailable when documents are private or still under design. This paper introduces SPECTRA, a re... | 2026-05-29 | [
"cs.IR",
"cs.AI"
] | [
-0.04572628438472748,
-0.04096699506044388,
-0.06567633897066116,
0.03722028806805611,
0.05589285492897034,
-0.05803373083472252,
0.04811190441250801,
0.033299803733825684,
-0.023447368294000626,
-0.0026170643977820873,
0.005105093587189913,
-0.04974149912595749,
0.015443546697497368,
-0.0... |
2605.31572 | nuReasoning: A Reasoning-Centric Dataset and Benchmark for Long-Tail Autonomous Driving | Reasoning is essential for autonomous driving (AD) in long-tail scenarios, where vehicles must apply commonsense knowledge, understand spatial relations, infer agent interactions, and make safe decisions. However, existing AD datasets and benchmarks mainly target perception, prediction, or planning, and provide limited... | 2026-05-29 | [
"cs.CV"
] | [
0.03537463769316673,
-0.07375802099704742,
0.00340764201246202,
0.04399022087454796,
0.10702964663505554,
0.0026707795914262533,
0.014219396747648716,
0.013118866831064224,
-0.012403972446918488,
-0.030719365924596786,
-0.03290785476565361,
-0.0917578712105751,
-0.03399203345179558,
0.1328... |
2605.31569 | A Datalog Framework for Conflict-Free Replicated Data Types | Distributed applications increasingly support local-first collaboration over shared data, allowing multiple users to perform updates concurrently without global coordination. Such collaboration requires careful design to capture the intended semantics of the concurrent interactions. We introduce a declarative framewo... | 2026-05-29 | [
"cs.DC",
"cs.DB",
"cs.LO",
"cs.PL"
] | [
0.007124417461454868,
0.007739552762359381,
0.02754146233201027,
-0.03434187173843384,
-0.0037611040752381086,
-0.10566596686840057,
-0.06175466999411583,
-0.04034848511219025,
-0.044673677533864975,
0.07649599760770798,
-0.04131381958723068,
-0.0339689701795578,
0.07490059733390808,
0.035... |
End of preview. Expand in Data Studio
Remyx Daily arXiv Pool
Last 30 days of arXiv submissions in selected CS categories, with abstracts pre-embedded using sentence-transformers/all-MiniLM-L6-v2 (384-dim).
Backs the candidate pre-filter step in the Remyx Paper Recommender Space. Pre-computed so the Space doesn't re-embed 500+ abstracts every cold start.
Schema
| Column | Type | Notes |
|---|---|---|
arxiv_id |
str | Without version suffix (e.g. 2605.23904) |
title |
str | One-line, newlines stripped |
abstract |
str | Original LaTeX-stripped abstract |
published_at |
str | YYYY-MM-DD |
categories |
list[str] | arXiv categories (e.g. ["cs.AI", "cs.CL"]) |
abstract_embedding |
list[float32] | 384-dim, NOT pre-normalized — normalize at use time |
Categories included
Current run: cs.AI, cs.LG, cs.CL.
Known limitation: this is AI/ML-heavy. Repos focused on adjacent domains (databases, distributed systems, computer vision) get weaker retrieval matches because their target papers aren't in the pool. Planned expansion: cs.CV, cs.DB, cs.SE, stat.ML in the next daily refresh.
Refresh cadence
Refreshed daily via scripts/update_arxiv_pool.py in the mhpd-dpo-training repo. Each refresh overwrites the parquet — older snapshots aren't retained.
Usage
from datasets import load_dataset
import numpy as np
ds = load_dataset("remyxai/arxiv_pool_daily", split="train")
embs = np.stack(ds["abstract_embedding"])
embs = embs / np.linalg.norm(embs, axis=1, keepdims=True) # normalize once
# Query: embed your text with the same model, then cosine top-k
from sentence_transformers import SentenceTransformer
m = SentenceTransformer("sentence-transformers/all-MiniLM-L6-v2")
q = m.encode("your team's domain summary", normalize_embeddings=True)
sims = embs @ q
top10 = np.argsort(-sims)[:10]
for i in top10:
print(ds[int(i)]["title"], sims[i])
- Downloads last month
- 22