title stringlengths 16 142 | paper_url stringlengths 43 45 | authors listlengths 1 45 | abstract large_stringlengths 346 1.74k ⌀ | anthology_id stringlengths 17 19 | doi stringlengths 29 31 | award stringclasses 0
values | arxiv_id stringlengths 10 10 ⌀ | arxiv_id_source stringclasses 2
values | embedding listlengths 768 768 |
|---|---|---|---|---|---|---|---|---|---|
Generative Knowledge Graph Construction: A Review | https://aclanthology.org/2022.emnlp-main.1/ | [
"Hongbin Ye",
"Ningyu Zhang",
"Hui Chen",
"Huajun Chen"
] | Generative Knowledge Graph Construction (KGC) refers to those methods that leverage the sequence-to-sequence framework for building knowledge graphs, which is flexible and can be adapted to widespread tasks. In this study, we summarize the recent compelling progress in generative knowledge graph construction. We presen... | 2022.emnlp-main.1 | 10.18653/v1/2022.emnlp-main.1 | null | 2210.12714 | title_snapshot | [
0.0123633136972785,
-0.045575108379125595,
-0.016206497326493263,
0.045819710940122604,
0.06410820782184601,
0.013794275000691414,
0.026586629450321198,
-0.006269113626331091,
0.023076577112078667,
-0.019237158820033073,
-0.01041442435234785,
0.022875601425766945,
-0.06343287229537964,
-0.... |
CDConv: A Benchmark for Contradiction Detection in Chinese Conversations | https://aclanthology.org/2022.emnlp-main.2/ | [
"Chujie Zheng",
"Jinfeng Zhou",
"Yinhe Zheng",
"Libiao Peng",
"Zhen Guo",
"Wenquan Wu",
"Zheng-Yu Niu",
"Hua Wu",
"Minlie Huang"
] | Dialogue contradiction is a critical issue in open-domain dialogue systems. The contextualization nature of conversations makes dialogue contradiction detection rather challenging. In this work, we propose a benchmark for Contradiction Detection in Chinese Conversations, namely CDConv. It contains 12K multi-turn conver... | 2022.emnlp-main.2 | 10.18653/v1/2022.emnlp-main.2 | null | 2210.08511 | title_snapshot | [
-0.019871758297085762,
-0.02299269661307335,
-0.06198424845933914,
0.0670379251241684,
0.009775049984455109,
0.019719291478395462,
0.024582037702202797,
0.04107850417494774,
0.01245494931936264,
-0.0337616428732872,
-0.02167338877916336,
0.06808315962553024,
-0.07097025960683823,
-0.008920... |
Transformer Feed-Forward Layers Build Predictions by Promoting Concepts in the Vocabulary Space | https://aclanthology.org/2022.emnlp-main.3/ | [
"Mor Geva",
"Avi Caciularu",
"Kevin Wang",
"Yoav Goldberg"
] | Transformer-based language models (LMs) are at the core of modern NLP, but their internal prediction construction process is opaque and largely not understood. In this work, we make a substantial step towards unveiling this underlying prediction process, by reverse-engineering the operation of the feed-forward network ... | 2022.emnlp-main.3 | 10.18653/v1/2022.emnlp-main.3 | null | 2203.14680 | title_snapshot | [
-0.01871603913605213,
-0.04845084249973297,
0.010049034841358662,
0.013567378744482994,
0.04691062495112419,
0.026628166437149048,
0.000994410365819931,
0.03404930979013443,
-0.018648013472557068,
-0.0001835728471633047,
-0.012403066270053387,
0.05271869897842407,
-0.04284710809588432,
0.0... |
Learning to Generate Question by Asking Question: A Primal-Dual Approach with Uncommon Word Generation | https://aclanthology.org/2022.emnlp-main.4/ | [
"Qifan Wang",
"Li Yang",
"Xiaojun Quan",
"Fuli Feng",
"Dongfang Liu",
"Zenglin Xu",
"Sinong Wang",
"Hao Ma"
] | Automatic question generation (AQG) is the task of generating a question from a given passage and an answer. Most existing AQG methods aim at encoding the passage and the answer to generate the question. However, limited work has focused on modeling the correlation between the target answer and the generated question. ... | 2022.emnlp-main.4 | 10.18653/v1/2022.emnlp-main.4 | null | null | null | [
0.005163456778973341,
-0.02170393243432045,
-0.006332279182970524,
0.0743880495429039,
0.034314706921577454,
0.022217467427253723,
0.02208876423537731,
-0.01935463584959507,
-0.017662733793258667,
-0.0037241228856146336,
-0.05207879841327667,
0.024379316717386246,
-0.06586947292089462,
-0.... |
Graph-based Model Generation for Few-Shot Relation Extraction | https://aclanthology.org/2022.emnlp-main.5/ | [
"Wanli Li",
"Tieyun Qian"
] | Few-shot relation extraction (FSRE) has been a challenging problem since it only has a handful of training instances. Existing models follow a ‘one-for-all’ scheme where one general large model performs all individual N-way-K-shot tasks in FSRE, which prevents the model from achieving the optimal point on each task. In... | 2022.emnlp-main.5 | 10.18653/v1/2022.emnlp-main.5 | null | null | null | [
-0.021709809079766273,
-0.020733091980218887,
0.022393692284822464,
0.04694047197699547,
0.025449521839618683,
0.025426553562283516,
0.03968403488397598,
0.0038554854691028595,
0.005226626992225647,
0.012254036962985992,
-0.021545113995671272,
0.042897891253232956,
-0.0807025209069252,
-0.... |
Backdoor Attacks in Federated Learning by Rare Embeddings and Gradient Ensembling | https://aclanthology.org/2022.emnlp-main.6/ | [
"Ki Yoon Yoo",
"Nojun Kwak"
] | Recent advances in federated learning have demonstrated its promising capability to learn on decentralized datasets. However, a considerable amount of work has raised concerns due to the potential risks of adversaries participating in the framework to poison the global model for an adversarial purpose. This paper inves... | 2022.emnlp-main.6 | 10.18653/v1/2022.emnlp-main.6 | null | 2204.14017 | title_snapshot | [
-0.013264535926282406,
-0.037951089441776276,
-0.016399402171373367,
0.06249839439988136,
0.050348956137895584,
-0.0026907729916274548,
0.058298125863075256,
-0.02364971861243248,
-0.009745129384100437,
-0.0065619610249996185,
-0.015341113321483135,
0.004254485946148634,
-0.03871282562613487... |
Generating Natural Language Proofs with Verifier-Guided Search | https://aclanthology.org/2022.emnlp-main.7/ | [
"Kaiyu Yang",
"Jia Deng",
"Danqi Chen"
] | Reasoning over natural language is a challenging problem in NLP. In this work, we focus on proof generation: Given a hypothesis and a set of supporting facts, the model generates a proof tree indicating how to derive the hypothesis from supporting facts. Compared to generating the entire proof in one shot, stepwise gen... | 2022.emnlp-main.7 | 10.18653/v1/2022.emnlp-main.7 | null | 2205.12443 | title_snapshot | [
-0.028177108615636826,
-0.022723481059074402,
0.0013840602478012443,
0.045554958283901215,
0.05605148896574974,
0.009514925070106983,
0.04621269553899765,
-0.0054807644337415695,
-0.022525936365127563,
0.004532165825366974,
-0.028395267203450203,
0.031233055517077446,
-0.028066454455256462,
... |
Toward Unifying Text Segmentation and Long Document Summarization | https://aclanthology.org/2022.emnlp-main.8/ | [
"Sangwoo Cho",
"Kaiqiang Song",
"Xiaoyang Wang",
"Fei Liu",
"Dong Yu"
] | Text segmentation is important for signaling a document’s structure. Without segmenting a long document into topically coherent sections, it is difficult for readers to comprehend the text, let alone find important information. The problem is only exacerbated by a lack of segmentation in transcripts of audio/video reco... | 2022.emnlp-main.8 | 10.18653/v1/2022.emnlp-main.8 | null | 2210.16422 | title_snapshot | [
-0.01743335835635662,
-0.03573649749159813,
-0.030049314722418785,
0.042844947427511215,
0.04377187043428421,
-0.019358914345502853,
0.02701422944664955,
0.022043032571673393,
-0.04048030078411102,
-0.029897579923272133,
-0.04095381498336792,
0.019117722287774086,
-0.04468042775988579,
0.0... |
The Geometry of Multilingual Language Model Representations | https://aclanthology.org/2022.emnlp-main.9/ | [
"Tyler A. Chang",
"Zhuowen Tu",
"Benjamin K. Bergen"
] | We assess how multilingual language models maintain a shared multilingual representation space while still encoding language-sensitive information in each language. Using XLM-R as a case study, we show that languages occupy similar linear subspaces after mean-centering, evaluated based on causal effects on language mod... | 2022.emnlp-main.9 | 10.18653/v1/2022.emnlp-main.9 | null | 2205.10964 | title_snapshot | [
-0.030053960159420967,
0.0018581716576591134,
0.007157590240240097,
0.0349552184343338,
0.008605152368545532,
0.033477526158094406,
0.021020136773586273,
0.025503942742943764,
-0.021079804748296738,
-0.004312705248594284,
-0.019833026453852654,
0.006340373307466507,
-0.05618475377559662,
0... |
Improving Complex Knowledge Base Question Answering via Question-to-Action and Question-to-Question Alignment | https://aclanthology.org/2022.emnlp-main.10/ | [
"Yechun Tang",
"Xiaoxia Cheng",
"Weiming Lu"
] | Complex knowledge base question answering can be achieved by converting questions into sequences of predefined actions. However, there is a significant semantic and structural gap between natural language and action sequences, which makes this conversion difficult. In this paper, we introduce an alignment-enhanced comp... | 2022.emnlp-main.10 | 10.18653/v1/2022.emnlp-main.10 | null | 2212.13036 | title_snapshot | [
0.04335656017065048,
-0.034838538616895676,
-0.014749865978956223,
0.049684230238199234,
0.029659409075975418,
-0.006380116101354361,
0.03497948870062828,
-0.009623625315725803,
0.004446659702807665,
-0.002606210531666875,
-0.028448961675167084,
0.03670521825551987,
-0.04559047520160675,
-... |
PAIR: Prompt-Aware margIn Ranking for Counselor Reflection Scoring in Motivational Interviewing | https://aclanthology.org/2022.emnlp-main.11/ | [
"Do June Min",
"Verónica Pérez-Rosas",
"Kenneth Resnicow",
"Rada Mihalcea"
] | Counselor reflection is a core verbal skill used by mental health counselors to express understanding and affirmation of the client’s experience and concerns. In this paper, we propose a system for the analysis of counselor reflections. Specifically, our system takes as input one dialog turn containing a client prompt ... | 2022.emnlp-main.11 | 10.18653/v1/2022.emnlp-main.11 | null | null | null | [
-0.014137841761112213,
-0.017055224627256393,
-0.002550646895542741,
0.005651138722896576,
0.026350470259785652,
0.015885118395090103,
0.041717417538166046,
-0.001169641618616879,
-0.014249465428292751,
-0.04058678075671196,
-0.02861647680401802,
0.021380649879574776,
-0.038272906094789505,
... |
Co-guiding Net: Achieving Mutual Guidances between Multiple Intent Detection and Slot Filling via Heterogeneous Semantics-Label Graphs | https://aclanthology.org/2022.emnlp-main.12/ | [
"Bowen Xing",
"Ivor Tsang"
] | Recent graph-based models for joint multiple intent detection and slot filling have obtained promising results through modeling the guidance from the prediction of intents to the decoding of slot filling.However, existing methods (1) only model the unidirectional guidance from intent to slot; (2) adopt homogeneous grap... | 2022.emnlp-main.12 | 10.18653/v1/2022.emnlp-main.12 | null | 2210.10375 | title_snapshot | [
0.027318619191646576,
-0.0381716787815094,
0.00035575765650719404,
0.007035544607788324,
0.0035446311812847853,
0.007597592659294605,
0.019223438575863838,
0.002949625952169299,
-0.01734653301537037,
-0.013567917048931122,
-0.034213341772556305,
0.01680619642138481,
-0.038584157824516296,
... |
The Importance of Being Parameters: An Intra-Distillation Method for Serious Gains | https://aclanthology.org/2022.emnlp-main.13/ | [
"Haoran Xu",
"Philipp Koehn",
"Kenton Murray"
] | Recent model pruning methods have demonstrated the ability to remove redundant parameters without sacrificing model performance. Common methods remove redundant parameters according to the parameter sensitivity, a gradient-based measure reflecting the contribution of the parameters. In this paper, however, we argue tha... | 2022.emnlp-main.13 | 10.18653/v1/2022.emnlp-main.13 | null | 2205.11416 | title_snapshot | [
-0.014542297460138798,
-0.010365376248955727,
0.015339731238782406,
0.019198058173060417,
0.024245191365480423,
0.036614369601011276,
0.04203169047832489,
-0.025869008153676987,
-0.004427093546837568,
-0.03069389797747135,
-0.035376593470573425,
0.00368137308396399,
-0.07536335289478302,
0... |
Interpreting Language Models with Contrastive Explanations | https://aclanthology.org/2022.emnlp-main.14/ | [
"Kayo Yin",
"Graham Neubig"
] | Model interpretability methods are often used to explain NLP model decisions on tasks such as text classification, where the output space is relatively small. However, when applied to language generation, where the output space often consists of tens of thousands of tokens, these methods are unable to provide informati... | 2022.emnlp-main.14 | 10.18653/v1/2022.emnlp-main.14 | null | 2202.10419 | title_snapshot | [
-0.03838304802775383,
-0.0036619033198803663,
-0.04533950239419937,
0.05765460804104805,
0.02566894143819809,
0.04340972751379013,
0.021292783319950104,
0.0026826136745512486,
-0.034771878272295,
0.02077990584075451,
-0.04198187217116356,
0.03695334121584892,
-0.03240802511572838,
0.014873... |
RankGen: Improving Text Generation with Large Ranking Models | https://aclanthology.org/2022.emnlp-main.15/ | [
"Kalpesh Krishna",
"Yapei Chang",
"John Wieting",
"Mohit Iyyer"
] | Given an input sequence (or prefix), modern language models often assign high probabilities to output sequences that are repetitive, incoherent, or irrelevant to the prefix; as such, model-generated text also contains such artifacts. To address these issues we present RankGen, a 1.2B parameter encoder model for English... | 2022.emnlp-main.15 | 10.18653/v1/2022.emnlp-main.15 | null | 2205.09726 | title_snapshot | [
-0.012055047787725925,
-0.03365040943026543,
-0.010837777517735958,
0.05849941074848175,
0.030945463106036186,
0.0027465198654681444,
0.02337287738919258,
0.004930229391902685,
-0.023189855739474297,
-0.02253340743482113,
-0.025661511346697807,
0.0365111343562603,
-0.039096664637327194,
-0... |
Learning a Grammar Inducer from Massive Uncurated Instructional Videos | https://aclanthology.org/2022.emnlp-main.16/ | [
"Songyang Zhang",
"Linfeng Song",
"Lifeng Jin",
"Haitao Mi",
"Kun Xu",
"Dong Yu",
"Jiebo Luo"
] | Video-aided grammar induction aims to leverage video information for finding more accurate syntactic grammars for accompanying text. While previous work focuses on building systems for inducing grammars on text that are well-aligned with video content, we investigate the scenario, in which text and video are only in lo... | 2022.emnlp-main.16 | 10.18653/v1/2022.emnlp-main.16 | null | 2210.12309 | title_snapshot | [
0.030704237520694733,
0.006556305568665266,
-0.010738952085375786,
0.04592248424887657,
0.0038583283312618732,
0.011819055303931236,
0.046335287392139435,
0.01388495322316885,
-0.0187698844820261,
-0.002364138141274452,
-0.029635289683938026,
0.043556783348321915,
-0.06422960013151169,
0.0... |
Normalized Contrastive Learning for Text-Video Retrieval | https://aclanthology.org/2022.emnlp-main.17/ | [
"Yookoon Park",
"Mahmoud Azab",
"Seungwhan Moon",
"Bo Xiong",
"Florian Metze",
"Gourab Kundu",
"Kirmani Ahmed"
] | Cross-modal contrastive learning has led the recent advances in multimodal retrieval with its simplicity and effectiveness. In this work, however, we reveal that cross-modal contrastive learning suffers from incorrect normalization of the sum retrieval probabilities of each text or video instance. Specifically, we show... | 2022.emnlp-main.17 | 10.18653/v1/2022.emnlp-main.17 | null | 2212.11790 | title_snapshot | [
0.007551232818514109,
-0.03444906696677208,
0.005968947894871235,
0.07184915244579315,
0.034237075597047806,
-0.02871735207736492,
0.02434852160513401,
0.012751723639667034,
-0.046783696860075,
0.005665292963385582,
-0.018055278807878494,
0.020790737122297287,
-0.05440094321966171,
-0.0018... |
Estimating Soft Labels for Out-of-Domain Intent Detection | https://aclanthology.org/2022.emnlp-main.18/ | [
"Hao Lang",
"Yinhe Zheng",
"Jian Sun",
"Fei Huang",
"Luo Si",
"Yongbin Li"
] | Out-of-Domain (OOD) intent detection is important for practical dialog systems. To alleviate the issue of lacking OOD training samples, some works propose synthesizing pseudo OOD samples and directly assigning one-hot OOD labels to these pseudo samples. However, these one-hot labels introduce noises to the training pro... | 2022.emnlp-main.18 | 10.18653/v1/2022.emnlp-main.18 | null | 2211.05561 | title_snapshot | [
-0.018225617706775665,
-0.023334309458732605,
-0.001487263012677431,
0.011416074819862843,
0.027778634801506996,
-0.00965478178113699,
0.045459307730197906,
0.016510121524333954,
-0.006819916423410177,
-0.0011137271067127585,
-0.054255448281764984,
0.039119902998209,
-0.07465104758739471,
... |
Multi-VQG: Generating Engaging Questions for Multiple Images | https://aclanthology.org/2022.emnlp-main.19/ | [
"Min-Hsuan Yeh",
"Vincent Chen",
"Ting-Hao Huang",
"Lun-Wei Ku"
] | Generating engaging content has drawn much recent attention in the NLP community. Asking questions is a natural way to respond to photos and promote awareness. However, most answers to questions in traditional question-answering (QA) datasets are factoids, which reduce individuals’ willingness to answer. Furthermore, t... | 2022.emnlp-main.19 | 10.18653/v1/2022.emnlp-main.19 | null | 2211.07441 | title_snapshot | [
0.024386664852499962,
-0.03834659606218338,
-0.00215104385279119,
0.059424277395009995,
0.01602719910442829,
0.014548740349709988,
0.03592664375901222,
0.011137071996927261,
-0.030298689380288124,
-0.009598408825695515,
-0.05796458572149277,
0.013669458217918873,
-0.07712794095277786,
-0.0... |
Tomayto, Tomahto. Beyond Token-level Answer Equivalence for Question Answering Evaluation | https://aclanthology.org/2022.emnlp-main.20/ | [
"Jannis Bulian",
"Christian Buck",
"Wojciech Gajewski",
"Benjamin Börschinger",
"Tal Schuster"
] | The predictions of question answering (QA) systems are typically evaluated against manually annotated finite sets of one or more answers. This leads to a coverage limitation that results in underestimating the true performance of systems, and is typically addressed by extending over exact match (EM) with predefined rul... | 2022.emnlp-main.20 | 10.18653/v1/2022.emnlp-main.20 | null | 2202.07654 | title_snapshot | [
-0.012636358849704266,
-0.03397221118211746,
-0.006317739840596914,
0.03993922099471092,
0.021069863811135292,
0.03444804251194,
0.0227081086486578,
-0.011790112592279911,
-0.0074936444871127605,
0.0014441357925534248,
-0.029139282181859016,
0.04813511669635773,
-0.049585625529289246,
-0.0... |
Non-Parametric Domain Adaptation for End-to-End Speech Translation | https://aclanthology.org/2022.emnlp-main.21/ | [
"Yichao Du",
"Weizhi Wang",
"Zhirui Zhang",
"Boxing Chen",
"Tong Xu",
"Jun Xie",
"Enhong Chen"
] | The end-to-end speech translation (E2E-ST) has received increasing attention due to the potential of its less error propagation, lower latency and fewer parameters. However, the effectiveness of neural-based approaches to this task is severely limited by the available training corpus, especially for domain adaptation w... | 2022.emnlp-main.21 | 10.18653/v1/2022.emnlp-main.21 | null | 2205.11211 | title_snapshot | [
-0.014826705679297447,
-0.009812208823859692,
-0.0037969667464494705,
0.036375921219587326,
0.032037097960710526,
0.07142282277345657,
0.04176589101552963,
-0.01829489879310131,
0.015825150534510612,
-0.0353257991373539,
-0.04264300689101219,
0.020666247233748436,
-0.043402452021837234,
0.... |
Prompting for Multimodal Hateful Meme Classification | https://aclanthology.org/2022.emnlp-main.22/ | [
"Rui Cao",
"Roy Ka-Wei Lee",
"Wen-Haw Chong",
"Jing Jiang"
] | Hateful meme classification is a challenging multimodal task that requires complex reasoning and contextual background knowledge. Ideally, we could leverage an explicit external knowledge base to supplement contextual and cultural information in hateful memes. However, there is no known explicit external knowledge base... | 2022.emnlp-main.22 | 10.18653/v1/2022.emnlp-main.22 | null | 2302.04156 | title_snapshot | [
0.024742823094129562,
-0.02047756500542164,
-0.01981900818645954,
0.03839531168341637,
0.004257731605321169,
0.040163855999708176,
0.0458650216460228,
0.010986777022480965,
-0.04117579385638237,
0.024409867823123932,
-0.025297611951828003,
0.030657202005386353,
-0.06313367187976837,
0.0063... |
Certified Error Control of Candidate Set Pruning for Two-Stage Relevance Ranking | https://aclanthology.org/2022.emnlp-main.23/ | [
"Minghan Li",
"Xinyu Zhang",
"Ji Xin",
"Hongyang Zhang",
"Jimmy Lin"
] | In information retrieval (IR), candidate set pruning has been commonly used to speed up two-stage relevance ranking. However, such an approach lacks accurate error control and often trades accuracy against computational efficiency in an empirical fashion, missing theoretical guarantees. In this paper, we propose the co... | 2022.emnlp-main.23 | 10.18653/v1/2022.emnlp-main.23 | null | 2205.09638 | title_snapshot | [
-0.024790586903691292,
-0.020151138305664062,
-0.013008099980652332,
0.04740576446056366,
0.05469154939055443,
0.006987590808421373,
0.019634820520877838,
-0.00911560095846653,
-0.0253546554595232,
-0.018083875998854637,
-0.04580610990524292,
0.040839314460754395,
-0.03295145928859711,
-0.... |
Linearizing Transformer with Key-Value Memory | https://aclanthology.org/2022.emnlp-main.24/ | [
"Yizhe Zhang",
"Deng Cai"
] | Efficient transformer variants with linear time complexity have been developed to mitigate the quadratic computational overhead of the vanilla transformer. Among them are low-rank projection methods such as Linformer and kernel-based Transformers. Despite their unique merits, they usually suffer from a performance drop... | 2022.emnlp-main.24 | 10.18653/v1/2022.emnlp-main.24 | null | 2203.12644 | title_snapshot | [
-0.046864621341228485,
-0.037929221987724304,
0.019871173426508904,
0.018390068784356117,
0.0289900079369545,
0.039425093680620193,
0.013720346614718437,
-0.00472353957593441,
-0.030350932851433754,
-0.018451111391186714,
0.011937330476939678,
0.006656541023403406,
-0.05941258743405342,
0.... |
Robustness of Fusion-based Multimodal Classifiers to Cross-Modal Content Dilutions | https://aclanthology.org/2022.emnlp-main.25/ | [
"Gaurav Verma",
"Vishwa Vinay",
"Ryan Rossi",
"Srijan Kumar"
] | As multimodal learning finds applications in a wide variety of high-stakes societal tasks, investigating their robustness becomes important. Existing work has focused on understanding the robustness of vision-and-language models to imperceptible variations on benchmark tasks. In this work, we investigate the robustness... | 2022.emnlp-main.25 | 10.18653/v1/2022.emnlp-main.25 | null | 2211.02646 | title_snapshot | [
0.012778419069945812,
-0.026703786104917526,
-0.008231938816606998,
0.058293625712394714,
0.007583064027130604,
0.008027225732803345,
0.025872960686683655,
0.025615360587835312,
-0.04216475412249565,
-0.016771506518125534,
-0.024060001596808434,
0.018488574773073196,
-0.06862202286720276,
... |
Translation between Molecules and Natural Language | https://aclanthology.org/2022.emnlp-main.26/ | [
"Carl Edwards",
"Tuan Lai",
"Kevin Ros",
"Garrett Honke",
"Kyunghyun Cho",
"Heng Ji"
] | We present MolT5 - a self-supervised learning framework for pretraining models on a vast amount of unlabeled natural language text and molecule strings. MolT5 allows for new, useful, and challenging analogs of traditional vision-language tasks, such as molecule captioning and text-based de novo molecule generation (alt... | 2022.emnlp-main.26 | 10.18653/v1/2022.emnlp-main.26 | null | 2204.11817 | title_snapshot | [
-0.008836772292852402,
-0.0013219621032476425,
0.015790298581123352,
0.05740145593881607,
0.03266188129782677,
-0.002599875908344984,
0.01622055284678936,
0.01895620860159397,
0.026921436190605164,
-0.00943422969430685,
-0.03945426270365715,
0.0574091374874115,
-0.06593282520771027,
-0.007... |
What Makes Instruction Learning Hard? An Investigation and a New Challenge in a Synthetic Environment | https://aclanthology.org/2022.emnlp-main.27/ | [
"Matthew Finlayson",
"Kyle Richardson",
"Ashish Sabharwal",
"Peter Clark"
] | The instruction learning paradigm—where a model learns to perform new tasks from task descriptions alone—has become popular in research on general-purpose models. The capabilities of large transformer models as instruction learners, however, remain poorly understood. We use a controlled synthetic environment to charact... | 2022.emnlp-main.27 | 10.18653/v1/2022.emnlp-main.27 | null | 2204.09148 | title_snapshot | [
-0.00950232520699501,
-0.037684813141822815,
-0.03001742623746395,
0.03281282261013985,
0.03717474639415741,
-0.005086516495794058,
0.04668116569519043,
0.021041037514805794,
0.00016324088210240006,
0.00518555473536253,
-0.03580287843942642,
0.059160053730010986,
-0.04383174329996109,
-0.0... |
Sentence-Incremental Neural Coreference Resolution | https://aclanthology.org/2022.emnlp-main.28/ | [
"Matt Grenander",
"Shay B. Cohen",
"Mark Steedman"
] | We propose a sentence-incremental neural coreference resolution system which incrementally builds clusters after marking mention boundaries in a shift-reduce method. The system is aimed at bridging two recent approaches at coreference resolution: (1) state-of-the-art non-incremental models that incur quadratic complexi... | 2022.emnlp-main.28 | 10.18653/v1/2022.emnlp-main.28 | null | 2305.16947 | title_snapshot | [
-0.028873592615127563,
-0.02896096184849739,
0.00006342918641166762,
0.04233802482485771,
0.04700974375009537,
0.05774447321891785,
-0.006644634995609522,
0.02061585709452629,
0.009066524915397167,
-0.013461711816489697,
-0.0035580950789153576,
0.02810116671025753,
-0.028726378455758095,
-... |
SNaC: Coherence Error Detection for Narrative Summarization | https://aclanthology.org/2022.emnlp-main.29/ | [
"Tanya Goyal",
"Junyi Jessy Li",
"Greg Durrett"
] | Progress in summarizing long texts is inhibited by the lack of appropriate evaluation frameworks. A long summary that appropriately covers the facets of that text must also present a coherent narrative, but current automatic and human evaluation methods fail to identify gaps in coherence. In this work, we introduce SNa... | 2022.emnlp-main.29 | 10.18653/v1/2022.emnlp-main.29 | null | 2205.09641 | title_snapshot | [
-0.014202236197888851,
-0.020580055192112923,
-0.016935721039772034,
0.022084981203079224,
0.038562651723623276,
-0.003466813126578927,
0.03130749240517616,
0.030139993876218796,
-0.035719580948352814,
-0.0035496738273650408,
-0.02626471035182476,
0.01991163194179535,
-0.045036863535642624,
... |
HydraSum: Disentangling Style Features in Text Summarization with Multi-Decoder Models | https://aclanthology.org/2022.emnlp-main.30/ | [
"Tanya Goyal",
"Nazneen Rajani",
"Wenhao Liu",
"Wojciech Kryscinski"
] | Summarization systems make numerous “decisions” about summary properties during inference, e.g. degree of copying, specificity and length of outputs, etc. However, these are implicitly encoded within model parameters and specific styles cannot be enforced. To address this, we introduce HydraSum, a new summarization arc... | 2022.emnlp-main.30 | 10.18653/v1/2022.emnlp-main.30 | null | 2110.04400 | title_judge | [
0.006095144432038069,
-0.021974407136440277,
-0.013330526649951935,
0.037533991038799286,
0.03285673260688782,
-0.006534770131111145,
0.013554966077208519,
0.013434769585728645,
-0.02598688006401062,
-0.022374141961336136,
-0.022810272872447968,
0.0482361875474453,
-0.05868922546505928,
-0... |
A Good Neighbor, A Found Treasure: Mining Treasured Neighbors for Knowledge Graph Entity Typing | https://aclanthology.org/2022.emnlp-main.31/ | [
"Zhuoran Jin",
"Pengfei Cao",
"Yubo Chen",
"Kang Liu",
"Jun Zhao"
] | The task of knowledge graph entity typing (KGET) aims to infer the missing types for entities in knowledge graphs. Some pioneering work has proved that neighbor information is very important for the task. However, existing methods only leverage the one-hop neighbor information of the central entity, ignoring the multi-... | 2022.emnlp-main.31 | 10.18653/v1/2022.emnlp-main.31 | null | null | null | [
-0.01212888304144144,
-0.037034496665000916,
0.025913219898939133,
0.04534885659813881,
0.05450166016817093,
0.009487371891736984,
0.03489496558904648,
-0.03799566999077797,
0.010234160348773003,
-0.03149206191301346,
-0.012289421632885933,
0.0030850444454699755,
-0.07019008696079254,
0.00... |
Guiding Neural Entity Alignment with Compatibility | https://aclanthology.org/2022.emnlp-main.32/ | [
"Bing Liu",
"Harrisen Scells",
"Wen Hua",
"Guido Zuccon",
"Genghong Zhao",
"Xia Zhang"
] | Entity Alignment (EA) aims to find equivalent entities between two Knowledge Graphs (KGs). While numerous neural EA models have been devised, they are mainly learned using labelled data only. In this work, we argue that different entities within one KG should have compatible counterparts in the other KG due to the pote... | 2022.emnlp-main.32 | 10.18653/v1/2022.emnlp-main.32 | null | 2211.15833 | title_snapshot | [
-0.007150718476623297,
-0.0022282209247350693,
0.0021593680139631033,
0.04404888302087784,
0.0472923181951046,
0.031208019703626633,
0.02695082500576973,
-0.001971068326383829,
-0.012065868824720383,
-0.030593236908316612,
-0.01907869428396225,
0.04087945818901062,
-0.060123227536678314,
-... |
InstructDial: Improving Zero and Few-shot Generalization in Dialogue through Instruction Tuning | https://aclanthology.org/2022.emnlp-main.33/ | [
"Prakhar Gupta",
"Cathy Jiao",
"Yi-Ting Yeh",
"Shikib Mehri",
"Maxine Eskenazi",
"Jeffrey Bigham"
] | Instruction tuning is an emergent paradigm in NLP wherein natural language instructions are leveraged with language models to induce zero-shot performance on unseen tasks. Dialogue is an especially interesting area in which to explore instruction tuning because dialogue systems perform multiple kinds of tasks related t... | 2022.emnlp-main.33 | 10.18653/v1/2022.emnlp-main.33 | null | 2205.12673 | title_snapshot | [
-0.024457085877656937,
-0.028089474886655807,
-0.041133906692266464,
0.06150103360414505,
0.013477877713739872,
0.04303867742419243,
0.05065510794520378,
0.01035352423787117,
-0.005445598158985376,
0.001973105361685157,
-0.057262174785137177,
0.05610775202512741,
-0.05649925023317337,
-0.0... |
Unsupervised Boundary-Aware Language Model Pretraining for Chinese Sequence Labeling | https://aclanthology.org/2022.emnlp-main.34/ | [
"Peijie Jiang",
"Dingkun Long",
"Yanzhao Zhang",
"Pengjun Xie",
"Meishan Zhang",
"Min Zhang"
] | Boundary information is critical for various Chinese language processing tasks, such as word segmentation, part-of-speech tagging, and named entity recognition. Previous studies usually resorted to the use of a high-quality external lexicon, where lexicon items can offer explicit boundary information. However, to ensur... | 2022.emnlp-main.34 | 10.18653/v1/2022.emnlp-main.34 | null | 2210.15231 | title_snapshot | [
-0.03467639535665512,
-0.014398115687072277,
-0.03838672861456871,
0.025126352906227112,
0.011357435956597328,
0.00489379744976759,
0.01929163746535778,
0.0031418204307556152,
0.02124379202723503,
-0.003606796497479081,
-0.054726701229810715,
0.0270406361669302,
-0.0506453663110733,
0.0018... |
RetroMAE: Pre-Training Retrieval-oriented Language Models Via Masked Auto-Encoder | https://aclanthology.org/2022.emnlp-main.35/ | [
"Shitao Xiao",
"Zheng Liu",
"Yingxia Shao",
"Zhao Cao"
] | Despite pre-training’s progress in many important NLP tasks, it remains to explore effective pre-training strategies for dense retrieval. In this paper, we propose RetroMAE, a new retrieval oriented pre-training paradigm based on Masked Auto-Encoder (MAE). RetroMAE is highlighted by three critical designs. 1) A novel M... | 2022.emnlp-main.35 | 10.18653/v1/2022.emnlp-main.35 | null | 2205.12035 | title_snapshot | [
-0.024605194106698036,
-0.0191965252161026,
-0.0012256457703188062,
0.02396846003830433,
0.028986085206270218,
0.02048686146736145,
0.03424091637134552,
0.0012422072468325496,
-0.024095889180898666,
-0.00820781197398901,
-0.039826899766922,
0.01911822147667408,
-0.04183291271328926,
0.0054... |
Aligning Recommendation and Conversation via Dual Imitation | https://aclanthology.org/2022.emnlp-main.36/ | [
"Jinfeng Zhou",
"Bo Wang",
"Minlie Huang",
"Dongming Zhao",
"Kun Huang",
"Ruifang He",
"Yuexian Hou"
] | Human conversations of recommendation naturally involve the shift of interests which can align the recommendation actions and conversation process to make accurate recommendations with rich explanations. However, existing conversational recommendation systems (CRS) ignore the advantage of user interest shift in connect... | 2022.emnlp-main.36 | 10.18653/v1/2022.emnlp-main.36 | null | 2211.02848 | title_snapshot | [
0.037984687834978104,
-0.027500979602336884,
-0.019262097775936127,
0.0376238115131855,
0.045149896293878555,
0.006739395204931498,
0.02302103117108345,
0.014301219023764133,
0.020266547799110413,
-0.03471900895237923,
-0.03470907360315323,
0.05066134035587311,
-0.046181004494428635,
-0.01... |
QRelScore: Better Evaluating Generated Questions with Deeper Understanding of Context-aware Relevance | https://aclanthology.org/2022.emnlp-main.37/ | [
"Xiaoqiang Wang",
"Bang Liu",
"Siliang Tang",
"Lingfei Wu"
] | Existing metrics for assessing question generation not only require costly human reference but also fail to take into account the input context of generation, rendering the lack of deep understanding of the relevance between the generated questions and input contexts. As a result, they may wrongly penalize a legitimate... | 2022.emnlp-main.37 | 10.18653/v1/2022.emnlp-main.37 | null | 2204.13921 | title_snapshot | [
-0.012172727845609188,
-0.03450661525130272,
-0.004729851149022579,
0.03863174095749855,
0.019901465624570847,
0.01761225052177906,
0.01547832041978836,
0.0022718191612511873,
-0.008223271928727627,
0.028731869533658028,
-0.046295590698719025,
0.06322279572486877,
-0.044264793395996094,
-0... |
Abstract Visual Reasoning with Tangram Shapes | https://aclanthology.org/2022.emnlp-main.38/ | [
"Anya Ji",
"Noriyuki Kojima",
"Noah Rush",
"Alane Suhr",
"Wai Keen Vong",
"Robert Hawkins",
"Yoav Artzi"
] | We introduce KiloGram, a resource for studying abstract visual reasoning in humans and machines. Drawing on the history of tangram puzzles as stimuli in cognitive science, we build a richly annotated dataset that, with >1k distinct stimuli, is orders of magnitude larger and more diverse than prior resources. It is both... | 2022.emnlp-main.38 | 10.18653/v1/2022.emnlp-main.38 | null | 2211.16492 | title_snapshot | [
-0.02510443888604641,
0.012582225725054741,
0.0016237021191045642,
0.027576643973588943,
0.02731899917125702,
0.009008776396512985,
0.015004795044660568,
0.03960713371634483,
-0.06580880284309387,
-0.023182949051260948,
-0.05251743271946907,
0.006726410239934921,
-0.06608384847640991,
-0.0... |
UnifiedSKG: Unifying and Multi-Tasking Structured Knowledge Grounding with Text-to-Text Language Models | https://aclanthology.org/2022.emnlp-main.39/ | [
"Tianbao Xie",
"Chen Henry Wu",
"Peng Shi",
"Ruiqi Zhong",
"Torsten Scholak",
"Michihiro Yasunaga",
"Chien-Sheng Wu",
"Ming Zhong",
"Pengcheng Yin",
"Sida I. Wang",
"Victor Zhong",
"Bailin Wang",
"Chengzu Li",
"Connor Boyle",
"Ansong Ni",
"Ziyu Yao",
"Dragomir Radev",
"Caiming Xion... | Structured knowledge grounding (SKG) leverages structured knowledge to complete user requests, such as semantic parsing over databases and question answering over knowledge bases. Since the inputs and outputs of SKG tasks are heterogeneous, they have been studied separately by different communities, which limits system... | 2022.emnlp-main.39 | 10.18653/v1/2022.emnlp-main.39 | null | 2201.05966 | title_snapshot | [
-0.002766565652564168,
-0.0377725251019001,
-0.005105257034301758,
0.03437410295009613,
0.03936128690838814,
0.007865565828979015,
0.029642164707183838,
0.0032048108987510204,
-0.00917744729667902,
-0.002246440388262272,
-0.0180495772510767,
0.029341604560613632,
-0.08402213454246521,
-0.0... |
Balanced Adversarial Training: Balancing Tradeoffs between Fickleness and Obstinacy in NLP Models | https://aclanthology.org/2022.emnlp-main.40/ | [
"Hannah Chen",
"Yangfeng Ji",
"David Evans"
] | Traditional (fickle) adversarial examples involve finding a small perturbation that does not change an input’s true label but confuses the classifier into outputting a different prediction. Conversely, obstinate adversarial examples occur when an adversary finds a small perturbation that preserves the classifier’s pred... | 2022.emnlp-main.40 | 10.18653/v1/2022.emnlp-main.40 | null | 2210.11498 | title_snapshot | [
-0.0046906727366149426,
-0.0373326800763607,
-0.02588736079633236,
0.04101342707872391,
0.011205571703612804,
0.020337413996458054,
0.03879665955901146,
0.000017827411284088157,
-0.016492415219545364,
-0.03220511972904205,
-0.02721690759062767,
0.018804898485541344,
-0.09447929263114929,
-... |
When Can Transformers Ground and Compose: Insights from Compositional Generalization Benchmarks | https://aclanthology.org/2022.emnlp-main.41/ | [
"Ankur Sikarwar",
"Arkil Patel",
"Navin Goyal"
] | Humans can reason compositionally whilst grounding language utterances to the real world. Recent benchmarks like ReaSCAN (Wu et al., 2021) use navigation tasks grounded in a grid world to assess whether neural models exhibit similar capabilities. In this work, we present a simple transformer-based model that outperform... | 2022.emnlp-main.41 | 10.18653/v1/2022.emnlp-main.41 | null | 2210.12786 | title_snapshot | [
-0.010227020829916,
-0.015982698649168015,
0.004060469567775726,
0.039901550859212875,
0.017072459682822227,
0.0012101155007258058,
0.015594670549035072,
0.041874390095472336,
-0.010214301757514477,
-0.013107677921652794,
-0.036285098642110825,
0.03592801094055176,
-0.062759168446064,
-0.0... |
Generative Language Models for Paragraph-Level Question Generation | https://aclanthology.org/2022.emnlp-main.42/ | [
"Asahi Ushio",
"Fernando Alva-Manchego",
"Jose Camacho-Collados"
] | Powerful generative models have led to recent progress in question generation (QG). However, it is difficult to measure advances in QG research since there are no standardized resources that allow a uniform comparison among approaches. In this paper, we introduce QG-Bench, a multilingual and multidomain benchmark for Q... | 2022.emnlp-main.42 | 10.18653/v1/2022.emnlp-main.42 | null | 2210.03992 | title_snapshot | [
-0.002156258327886462,
-0.033363424241542816,
-0.01587647758424282,
0.06190747395157814,
0.01736103929579258,
0.036125652492046356,
0.012877361848950386,
0.019880300387740135,
-0.007860560901463032,
0.0046382793225348,
-0.02950313687324524,
0.025157039985060692,
-0.05646365508437157,
-0.01... |
A Unified Encoder-Decoder Framework with Entity Memory | https://aclanthology.org/2022.emnlp-main.43/ | [
"Zhihan Zhang",
"Wenhao Yu",
"Chenguang Zhu",
"Meng Jiang"
] | Entities, as important carriers of real-world knowledge, play a key role in many NLP tasks.We focus on incorporating entity knowledge into an encoder-decoder framework for informative text generation. Existing approaches tried to index, retrieve, and read external documents as evidence, but they suffered from a large c... | 2022.emnlp-main.43 | 10.18653/v1/2022.emnlp-main.43 | null | 2210.03273 | title_snapshot | [
-0.015188711695373058,
-0.01699487492442131,
0.000881574465893209,
0.08455484360456467,
0.032930467277765274,
0.00353295193053782,
0.010098211467266083,
0.027734510600566864,
-0.031793415546417236,
-0.004342824220657349,
-0.03542478010058403,
0.029790053144097328,
-0.05142894759774208,
0.0... |
Segmenting Numerical Substitution Ciphers | https://aclanthology.org/2022.emnlp-main.44/ | [
"Nada Aldarrab",
"Jonathan May"
] | Deciphering historical substitution ciphers is a challenging problem. Example problems that have been previously studied include detecting cipher type, detecting plaintext language, and acquiring the substitution key for segmented ciphers. However, attacking unsegmented ciphers is still a challenging task. Segmentation... | 2022.emnlp-main.44 | 10.18653/v1/2022.emnlp-main.44 | null | 2205.12527 | title_snapshot | [
-0.03930918127298355,
0.017749348655343056,
-0.0447230227291584,
0.057420436292886734,
0.056824591010808945,
0.032239414751529694,
0.055073630064725876,
0.006139932665973902,
-0.03699479624629021,
-0.03739788755774498,
-0.01647038385272026,
-0.010062327608466148,
-0.0435849130153656,
0.017... |
Crossmodal-3600: A Massively Multilingual Multimodal Evaluation Dataset | https://aclanthology.org/2022.emnlp-main.45/ | [
"Ashish V. Thapliyal",
"Jordi Pont Tuset",
"Xi Chen",
"Radu Soricut"
] | Research in massively multilingual image captioning has been severely hampered by a lack of high-quality evaluation datasets. In this paper we present the Crossmodal-3600 dataset (XM3600 in short), a geographically diverse set of 3600 images annotated with human-generated reference captions in 36 languages. The images ... | 2022.emnlp-main.45 | 10.18653/v1/2022.emnlp-main.45 | null | 2205.12522 | title_snapshot | [
0.0009967355290427804,
-0.037393324077129364,
0.001000459655188024,
0.05913935601711273,
0.015433800406754017,
0.01131383702158928,
0.02981957420706749,
0.04957221448421478,
-0.020284799858927727,
-0.009579818695783615,
-0.06360064446926117,
0.03489852324128151,
-0.06989986449480057,
-0.01... |
ReSel: N-ary Relation Extraction from Scientific Text and Tables by Learning to Retrieve and Select | https://aclanthology.org/2022.emnlp-main.46/ | [
"Yuchen Zhuang",
"Yinghao Li",
"Junyang Zhang",
"Yue Yu",
"Yingjun Mou",
"Xiang Chen",
"Le Song",
"Chao Zhang"
] | We study the problem of extracting N-ary relation tuples from scientific articles. This task is challenging because the target knowledge tuples can reside in multiple parts and modalities of the document. Our proposed method ReSel decomposes this task into a two-stage procedure that first retrieves the most relevant pa... | 2022.emnlp-main.46 | 10.18653/v1/2022.emnlp-main.46 | null | 2210.14427 | title_snapshot | [
-0.01666698232293129,
-0.010942681692540646,
0.013529844582080841,
0.060520563274621964,
0.05096035450696945,
-0.008598757907748222,
0.005145031958818436,
0.00008358967897947878,
-0.0017883768305182457,
-0.004872170742601156,
-0.02569636143743992,
0.05796707421541214,
-0.06277643144130707,
... |
GammaE: Gamma Embeddings for Logical Queries on Knowledge Graphs | https://aclanthology.org/2022.emnlp-main.47/ | [
"Dong Yang",
"Peijun Qing",
"Yang Li",
"Haonan Lu",
"Xiaodong Lin"
] | Embedding knowledge graphs (KGs) for multi-hop logical reasoning is a challenging problem due to massive and complicated structures in many KGs. Recently, many promising works projected entities and queries into a geometric space to efficiently find answers. However, it remains challenging to model the negation and uni... | 2022.emnlp-main.47 | 10.18653/v1/2022.emnlp-main.47 | null | 2210.15578 | title_snapshot | [
0.006140833720564842,
-0.021926479414105415,
0.0031908308155834675,
0.04905198886990547,
0.05242254585027695,
-0.008156687021255493,
0.04091929271817207,
-0.03519577533006668,
0.010424124076962471,
-0.020435648038983345,
-0.021199293434619904,
0.031407564878463745,
-0.0795602798461914,
0.0... |
Reasoning Like Program Executors | https://aclanthology.org/2022.emnlp-main.48/ | [
"Xinyu Pi",
"Qian Liu",
"Bei Chen",
"Morteza Ziyadi",
"Zeqi Lin",
"Qiang Fu",
"Yan Gao",
"Jian-Guang Lou",
"Weizhu Chen"
] | Reasoning over natural language is a long-standing goal for the research community. However, studies have shown that existing language models are inadequate in reasoning. To address the issue, we present POET, a novel reasoning pre-training paradigm. Through pre-training language models with programs and their executio... | 2022.emnlp-main.48 | 10.18653/v1/2022.emnlp-main.48 | null | 2201.11473 | title_snapshot | [
-0.03261170536279678,
-0.03045770339667797,
-0.02671206183731556,
0.04405174031853676,
0.055911894887685776,
0.017082080245018005,
0.018193379044532776,
0.0008253221749328077,
-0.029090311378240585,
-0.026513880118727684,
-0.019492654129862785,
0.05475517362356186,
-0.06481240689754486,
-0... |
SEM-F1: an Automatic Way for Semantic Evaluation of Multi-Narrative Overlap Summaries at Scale | https://aclanthology.org/2022.emnlp-main.49/ | [
"Naman Bansal",
"Mousumi Akter",
"Shubhra Kanti Karmaker Santu"
] | Recent work has introduced an important yet relatively under-explored NLP task called Semantic Overlap Summarization (SOS) that entails generating a summary from multiple alternative narratives which conveys the common information provided by those narratives. Previous work also published a benchmark dataset for this t... | 2022.emnlp-main.49 | 10.18653/v1/2022.emnlp-main.49 | null | 2201.05294 | title_judge | [
-0.025213029235601425,
-0.034758806228637695,
-0.020106540992856026,
0.03253035992383957,
0.03781379014253616,
0.005805076099932194,
0.027903128415346146,
0.012223981320858002,
-0.00567376846447587,
-0.005720443092286587,
-0.029605116695165634,
0.029914377257227898,
-0.05317366123199463,
-... |
Inducer-tuning: Connecting Prefix-tuning and Adapter-tuning | https://aclanthology.org/2022.emnlp-main.50/ | [
"Yifan Chen",
"Devamanyu Hazarika",
"Mahdi Namazifar",
"Yang Liu",
"Di Jin",
"Dilek Hakkani-Tur"
] | Prefix-tuning, or more generally continuous prompt tuning, has become an essential paradigm of parameter-efficient transfer learning. Using a large pre-trained language model (PLM), prefix-tuning can obtain strong performance by training only a small portion of parameters. In this paper, we propose to understand and fu... | 2022.emnlp-main.50 | 10.18653/v1/2022.emnlp-main.50 | null | 2210.14469 | title_snapshot | [
-0.0348319448530674,
-0.04517931118607521,
-0.00558058638125658,
0.030942406505346298,
0.037896934896707535,
0.052811380475759506,
0.02195148728787899,
-0.01338131818920374,
-0.025731168687343597,
-0.01494584884494543,
-0.0153976920992136,
0.056187666952610016,
-0.028842071071267128,
0.000... |
DocInfer: Document-level Natural Language Inference using Optimal Evidence Selection | https://aclanthology.org/2022.emnlp-main.51/ | [
"Puneet Mathur",
"Gautam Kunapuli",
"Riyaz Bhat",
"Manish Shrivastava",
"Dinesh Manocha",
"Maneesh Singh"
] | We present DocInfer - a novel, end-to-end Document-level Natural Language Inference model that builds a hierarchical document graph enriched through inter-sentence relations (topical, entity-based, concept-based), performs paragraph pruning using the novel SubGraph Pooling layer, followed by optimal evidence selection ... | 2022.emnlp-main.51 | 10.18653/v1/2022.emnlp-main.51 | null | null | null | [
-0.013469837605953217,
-0.046289220452308655,
-0.016962101683020592,
0.07167612761259079,
0.03263888880610466,
-0.009679736569523811,
0.02851281501352787,
-0.018633654341101646,
-0.004089852795004845,
-0.006213916465640068,
-0.007344549056142569,
0.08025062084197998,
-0.04101352393627167,
... |
LightEA: A Scalable, Robust, and Interpretable Entity Alignment Framework via Three-view Label Propagation | https://aclanthology.org/2022.emnlp-main.52/ | [
"Xin Mao",
"Wenting Wang",
"Yuanbin Wu",
"Man Lan"
] | Entity Alignment (EA) aims to find equivalent entity pairs between KGs, which is the core step to bridging and integrating multi-source KGs. In this paper, we argue that existing complex EA methods inevitably inherit the inborn defects from their neural network lineage: poor interpretability and weak scalability. Inspi... | 2022.emnlp-main.52 | 10.18653/v1/2022.emnlp-main.52 | null | 2210.10436 | title_snapshot | [
0.006293379236012697,
0.002098845550790429,
-0.0026795961894094944,
0.03397396579384804,
0.019332991912961006,
0.016462862491607666,
0.02400553785264492,
-0.012064994312822819,
-0.023256823420524597,
-0.019996367394924164,
0.0030481081921607256,
0.013192142359912395,
-0.0689157024025917,
0... |
Metric-guided Distillation: Distilling Knowledge from the Metric to Ranker and Retriever for Generative Commonsense Reasoning | https://aclanthology.org/2022.emnlp-main.53/ | [
"Xingwei He",
"Yeyun Gong",
"A-Long Jin",
"Weizhen Qi",
"Hang Zhang",
"Jian Jiao",
"Bartuer Zhou",
"Biao Cheng",
"Sm Yiu",
"Nan Duan"
] | Commonsense generation aims to generate a realistic sentence describing a daily scene under the given concepts, which is very challenging, since it requires models to have relational reasoning and compositional generalization capabilities. Previous work focuses on retrieving prototype sentences for the provided concept... | 2022.emnlp-main.53 | 10.18653/v1/2022.emnlp-main.53 | null | 2210.11708 | title_snapshot | [
-0.0146102886646986,
-0.019215403124690056,
-0.005544244311749935,
0.055290039628744125,
0.03985988721251488,
-0.01046499703079462,
0.028063533827662468,
-0.005127900280058384,
-0.009568478912115097,
0.011855004355311394,
-0.033794816583395004,
0.01080170925706625,
-0.05203844979405403,
0.... |
Efficient Document Retrieval by End-to-End Refining and Quantizing BERT Embedding with Contrastive Product Quantization | https://aclanthology.org/2022.emnlp-main.54/ | [
"Zexuan Qiu",
"Qinliang Su",
"Jianxing Yu",
"Shijing Si"
] | Efficient document retrieval heavily relies on the technique of semantic hashing, which learns a binary code for every document and employs Hamming distance to evaluate document distances. However, existing semantic hashing methods are mostly established on outdated TFIDF features, which obviously do not contain lots o... | 2022.emnlp-main.54 | 10.18653/v1/2022.emnlp-main.54 | null | 2210.17170 | title_snapshot | [
-0.017177026718854904,
-0.03666511923074722,
-0.025131521746516228,
0.06483210623264313,
0.03238192945718765,
-0.012982397340238094,
-0.009298300370573997,
-0.01647799462080002,
-0.003644680604338646,
-0.020076675340533257,
-0.02629099227488041,
0.0028676935471594334,
-0.04519400745630264,
... |
Curriculum Knowledge Distillation for Emoji-supervised Cross-lingual Sentiment Analysis | https://aclanthology.org/2022.emnlp-main.55/ | [
"Jianyang Zhang",
"Tao Liang",
"Mingyang Wan",
"Guowu Yang",
"Fengmao Lv"
] | Existing sentiment analysis models have achieved great advances with the help of sufficient sentiment annotations. Unfortunately, many languages do not have sufficient sentiment corpus. To this end, recent studies have proposed cross-lingual sentiment analysis to transfer sentiment analysis models from resource-rich la... | 2022.emnlp-main.55 | 10.18653/v1/2022.emnlp-main.55 | null | null | null | [
0.033408354967832565,
-0.022427039220929146,
-0.009482908062636852,
0.051347516477108,
0.030672332271933556,
0.011679315939545631,
0.03052506409585476,
0.03728729858994484,
-0.0014118081890046597,
0.01755097508430481,
-0.053331565111875534,
0.010623098351061344,
-0.047238629311323166,
-0.0... |
Correctable-DST: Mitigating Historical Context Mismatch between Training and Inference for Improved Dialogue State Tracking | https://aclanthology.org/2022.emnlp-main.56/ | [
"Hongyan Xie",
"Haoxiang Su",
"Shuangyong Song",
"Hao Huang",
"Bo Zou",
"Kun Deng",
"Jianghua Lin",
"Zhihui Zhang",
"Xiaodong He"
] | Recently proposed dialogue state tracking (DST) approaches predict the dialogue state of a target turn sequentially based on the previous dialogue state. During the training time, the ground-truth previous dialogue state is utilized as the historical context. However, only the previously predicted dialogue state can be... | 2022.emnlp-main.56 | 10.18653/v1/2022.emnlp-main.56 | null | null | null | [
-0.040819838643074036,
-0.000931441958528012,
-0.004636486526578665,
0.037320576608181,
0.06096531078219414,
0.031603675335645676,
0.04361557960510254,
0.030321629717946053,
-0.008839532732963562,
-0.04871141538023949,
-0.024319684132933617,
0.07099995017051697,
-0.04223297908902168,
-0.02... |
DropMix: A Textual Data Augmentation Combining Dropout with Mixup | https://aclanthology.org/2022.emnlp-main.57/ | [
"Fanshuang Kong",
"Richong Zhang",
"Xiaohui Guo",
"Samuel Mensah",
"Yongyi Mao"
] | Overfitting is a notorious problem when there is insufficient data to train deep neural networks in machine learning tasks. Data augmentation regularization methods such as Dropout, Mixup, and their enhanced variants are effective and prevalent, and achieve promising performance to overcome overfitting. However, in tex... | 2022.emnlp-main.57 | 10.18653/v1/2022.emnlp-main.57 | null | null | null | [
0.009388219565153122,
-0.057512737810611725,
-0.025523487478494644,
0.05772867798805237,
0.03353748470544815,
0.03256276994943619,
0.007197360508143902,
0.03445545956492424,
-0.02405945025384426,
-0.038599081337451935,
-0.046381957828998566,
0.018454883247613907,
-0.06938594579696655,
-0.0... |
Cross-document Event Coreference Search: Task, Dataset and Modeling | https://aclanthology.org/2022.emnlp-main.58/ | [
"Alon Eirew",
"Avi Caciularu",
"Ido Dagan"
] | The task of Cross-document Coreference Resolution has been traditionally formulated as requiring to identify all coreference links across a given set of documents. We propose an appealing, and often more applicable, complementary set up for the task – Cross-document Coreference Search, focusing in this paper on event c... | 2022.emnlp-main.58 | 10.18653/v1/2022.emnlp-main.58 | null | 2210.12654 | title_snapshot | [
-0.02253754809498787,
-0.02429656684398651,
-0.012482097372412682,
0.03312024101614952,
0.03521728515625,
-0.021234959363937378,
0.012298477813601494,
0.009550662711262703,
0.0034061926417052746,
-0.017103109508752823,
-0.028252264484763145,
0.018378568813204765,
-0.044542860239744186,
-0.... |
VIRT: Improving Representation-based Text Matching via Virtual Interaction | https://aclanthology.org/2022.emnlp-main.59/ | [
"Dan Li",
"Yang Yang",
"Hongyin Tang",
"Jiahao Liu",
"Qifan Wang",
"Jingang Wang",
"Tong Xu",
"Wei Wu",
"Enhong Chen"
] | Text matching is a fundamental research problem in natural language understanding. Interaction-based approaches treat the text pair as a single sequence and encode it through cross encoders, while representation-based models encode the text pair independently with siamese or dual encoders. Interaction-based models requ... | 2022.emnlp-main.59 | 10.18653/v1/2022.emnlp-main.59 | null | 2112.04195 | title_judge | [
0.02195035107433796,
-0.0027803201228380203,
-0.005238536279648542,
0.07218457013368607,
0.03451211005449295,
0.016671543940901756,
0.038327716290950775,
0.017316022887825966,
0.006381572689861059,
-0.0218735933303833,
-0.05480716750025749,
0.031939469277858734,
-0.07412862032651901,
-0.00... |
MAVEN-ERE: A Unified Large-scale Dataset for Event Coreference, Temporal, Causal, and Subevent Relation Extraction | https://aclanthology.org/2022.emnlp-main.60/ | [
"Xiaozhi Wang",
"Yulin Chen",
"Ning Ding",
"Hao Peng",
"Zimu Wang",
"Yankai Lin",
"Xu Han",
"Lei Hou",
"Juanzi Li",
"Zhiyuan Liu",
"Peng Li",
"Jie Zhou"
] | The diverse relationships among real-world events, including coreference, temporal, causal, and subevent relations, are fundamental to understanding natural languages. However, two drawbacks of existing datasets limit event relation extraction (ERE) tasks: (1) Small scale. Due to the annotation complexity, the data sca... | 2022.emnlp-main.60 | 10.18653/v1/2022.emnlp-main.60 | null | 2211.07342 | title_snapshot | [
-0.028227616101503372,
-0.03203488513827324,
-0.016069984063506126,
0.02832472324371338,
0.029428256675601006,
0.026609931141138077,
0.02094901353120804,
0.008630895987153053,
-0.03251102939248085,
-0.0027177154552191496,
-0.037476107478141785,
0.04705560952425003,
-0.07298701256513596,
0.... |
Entity Extraction in Low Resource Domains with Selective Pre-training of Large Language Models | https://aclanthology.org/2022.emnlp-main.61/ | [
"Aniruddha Mahapatra",
"Sharmila Reddy Nangi",
"Aparna Garimella",
"Anandhavelu N"
] | Transformer-based language models trained on large natural language corpora have been very useful in downstream entity extraction tasks. However, they often result in poor performances when applied to domains that are different from those they are pretrained on. Continued pretraining using unlabeled data from target do... | 2022.emnlp-main.61 | 10.18653/v1/2022.emnlp-main.61 | null | null | null | [
-0.028638800606131554,
-0.020053917542099953,
0.0036430389154702425,
0.05666082352399826,
0.04376581311225891,
0.012401796877384186,
0.009902018122375011,
0.008018004707992077,
0.0025145672261714935,
0.015493604354560375,
-0.05283863842487335,
0.0474972166121006,
-0.06853166222572327,
0.01... |
How Large Language Models are Transforming Machine-Paraphrase Plagiarism | https://aclanthology.org/2022.emnlp-main.62/ | [
"Jan Philip Wahle",
"Terry Ruas",
"Frederic Kirstein",
"Bela Gipp"
] | The recent success of large language models for text generation poses a severe threat to academic integrity, as plagiarists can generate realistic paraphrases indistinguishable from original work.However, the role of large autoregressive models in generating machine-paraphrased plagiarism and their detection is still i... | 2022.emnlp-main.62 | 10.18653/v1/2022.emnlp-main.62 | null | null | null | [
0.0012139256577938795,
-0.017064452171325684,
-0.03674287721514702,
0.05086558684706688,
0.054801493883132935,
0.01555327232927084,
0.03466956689953804,
0.03764815255999565,
-0.020999440923333168,
-0.00561971589922905,
-0.020677654072642326,
0.030721649527549744,
-0.04401993006467819,
0.01... |
M2D2: A Massively Multi-Domain Language Modeling Dataset | https://aclanthology.org/2022.emnlp-main.63/ | [
"Machel Reid",
"Victor Zhong",
"Suchin Gururangan",
"Luke Zettlemoyer"
] | We present M2D2, a fine-grained, massively multi-domain corpus for studying domain adaptation in language models (LMs). M2D2 consists of 8.5B tokens and spans 145 domains extracted from Wikipedia and Semantic Scholar. Using ontologies derived from Wikipedia and ArXiv categories, we organize the domains in each data sou... | 2022.emnlp-main.63 | 10.18653/v1/2022.emnlp-main.63 | null | 2210.07370 | title_snapshot | [
-0.02971619740128517,
-0.01234806515276432,
-0.011974874883890152,
0.04661819711327553,
0.04067855328321457,
0.029956644400954247,
0.0278784092515707,
0.007905484177172184,
-0.012514163739979267,
0.0038937439676374197,
-0.021319763734936714,
0.030117057263851166,
-0.053309231996536255,
0.0... |
“Will You Find These Shortcuts?” A Protocol for Evaluating the Faithfulness of Input Salience Methods for Text Classification | https://aclanthology.org/2022.emnlp-main.64/ | [
"Jasmijn Bastings",
"Sebastian Ebert",
"Polina Zablotskaia",
"Anders Sandholm",
"Katja Filippova"
] | Feature attribution a.k.a. input salience methods which assign an importance score to a feature are abundant but may produce surprisingly different results for the same model on the same input. While differences are expected if disparate definitions of importance are assumed, most methods claim to provide faithful attr... | 2022.emnlp-main.64 | 10.18653/v1/2022.emnlp-main.64 | null | 2111.07367 | title_snapshot | [
-0.04435106739401817,
-0.02899307571351528,
-0.017230292782187462,
0.04988708347082138,
-0.0007104304386302829,
0.03712953254580498,
0.048059988766908646,
0.00958674680441618,
-0.006735025439411402,
-0.01462895330041647,
-0.050249338150024414,
0.0640702024102211,
-0.03910429775714874,
-0.0... |
Information-Transport-based Policy for Simultaneous Translation | https://aclanthology.org/2022.emnlp-main.65/ | [
"Shaolei Zhang",
"Yang Feng"
] | Simultaneous translation (ST) outputs translation while receiving the source inputs, and hence requires a policy to determine whether to translate a target token or wait for the next source token. The major challenge of ST is that each target token can only be translated based on the current received source tokens, whe... | 2022.emnlp-main.65 | 10.18653/v1/2022.emnlp-main.65 | null | 2210.12357 | title_snapshot | [
-0.041581589728593826,
-0.021669790148735046,
0.009837391786277294,
0.028440942987799644,
0.01615644060075283,
0.02468503639101982,
0.023601911962032318,
0.02347203530371189,
-0.017555251717567444,
-0.0343136228621006,
-0.0288548581302166,
0.009928644634783268,
-0.03658783435821533,
-0.005... |
Learning to Adapt to Low-Resource Paraphrase Generation | https://aclanthology.org/2022.emnlp-main.66/ | [
"Zhigen Li",
"Yanmeng Wang",
"Rizhao Fan",
"Ye Wang",
"Jianfeng Li",
"Shaojun Wang"
] | Paraphrase generation is a longstanding NLP task and achieves great success with the aid of large corpora. However, transferring a paraphrasing model to another domain encounters the problem of domain shifting especially when the data is sparse. At the same time, widely using large pre-trained language models (PLMs) fa... | 2022.emnlp-main.66 | 10.18653/v1/2022.emnlp-main.66 | null | 2412.17111 | title_snapshot | [
0.0021880066487938166,
-0.018727999180555344,
-0.011239593848586082,
0.056592050939798355,
0.062195371836423874,
0.05129753053188324,
0.00580973643809557,
0.006061075255274773,
-0.013485102914273739,
-0.00041201620479114354,
-0.03122590109705925,
0.044098637998104095,
-0.06509978324174881,
... |
A Distributional Lens for Multi-Aspect Controllable Text Generation | https://aclanthology.org/2022.emnlp-main.67/ | [
"Yuxuan Gu",
"Xiaocheng Feng",
"Sicheng Ma",
"Lingyuan Zhang",
"Heng Gong",
"Bing Qin"
] | Multi-aspect controllable text generation is a more challenging and practical task than single-aspect control. Existing methods achieve complex multi-aspect control by fusing multiple controllers learned from single-aspect, but suffer from attribute degeneration caused by the mutual interference of these controllers. T... | 2022.emnlp-main.67 | 10.18653/v1/2022.emnlp-main.67 | null | 2210.02889 | title_snapshot | [
-0.02301127277314663,
-0.003510642796754837,
-0.00821099802851677,
0.0643022283911705,
0.02546488307416439,
0.026062237098813057,
0.021751193329691887,
0.003859817748889327,
0.004663245286792517,
-0.022410664707422256,
-0.023939302191138268,
0.028569981455802917,
-0.07015945762395859,
-0.0... |
ELMER: A Non-Autoregressive Pre-trained Language Model for Efficient and Effective Text Generation | https://aclanthology.org/2022.emnlp-main.68/ | [
"Junyi Li",
"Tianyi Tang",
"Wayne Xin Zhao",
"Jian-Yun Nie",
"Ji-Rong Wen"
] | We study the text generation task under the approach of pre-trained language models (PLMs). Typically, an auto-regressive (AR) method is adopted for generating texts in a token-by-token manner. Despite many advantages of AR generation, it usually suffers from inefficient inference. Therefore, non-autoregressive (NAR) m... | 2022.emnlp-main.68 | 10.18653/v1/2022.emnlp-main.68 | null | 2210.13304 | title_snapshot | [
-0.009264517575502396,
-0.018410826101899147,
-0.021702775731682777,
0.0370849072933197,
0.04278077557682991,
0.027127353474497795,
0.020260583609342575,
0.002573666861280799,
-0.026620393618941307,
-0.013935931026935577,
-0.02898004837334156,
0.00621828343719244,
-0.05729253217577934,
0.0... |
Multilingual Relation Classification via Efficient and Effective Prompting | https://aclanthology.org/2022.emnlp-main.69/ | [
"Yuxuan Chen",
"David Harbecke",
"Leonhard Hennig"
] | Prompting pre-trained language models has achieved impressive performance on various NLP tasks, especially in low data regimes. Despite the success of prompting in monolingual settings, applying prompt-based methods in multilingual scenarios has been limited to a narrow set of tasks, due to the high cost of handcraftin... | 2022.emnlp-main.69 | 10.18653/v1/2022.emnlp-main.69 | null | 2210.13838 | title_snapshot | [
-0.019243456423282623,
-0.02964332327246666,
0.021940505132079124,
0.06210418418049812,
0.009779415093362331,
0.018650826066732407,
-0.00789167545735836,
-0.0005131645011715591,
-0.004544359166175127,
0.03347795829176903,
-0.04345613718032837,
0.05498731508851051,
-0.062238939106464386,
-0... |
Topic-Regularized Authorship Representation Learning | https://aclanthology.org/2022.emnlp-main.70/ | [
"Jitkapat Sawatphol",
"Nonthakit Chaiwong",
"Can Udomcharoenchaikit",
"Sarana Nutanong"
] | Authorship attribution is a task that aims to identify the author of a given piece of writing. We aim to develop a generalized solution that can handle a large number of texts from authors and topics unavailable in training data. Previous studies have proposed strategies to address only either unseen authors or unseen ... | 2022.emnlp-main.70 | 10.18653/v1/2022.emnlp-main.70 | null | null | null | [
0.01069282554090023,
-0.021144984290003777,
-0.014980112202465534,
0.04332534596323967,
0.05919547751545906,
-0.015827201306819916,
0.010960608720779419,
0.002360178390517831,
-0.031043345108628273,
-0.019751491025090218,
-0.01028071716427803,
0.01783222332596779,
-0.06894433498382568,
-0.... |
Fine-grained Contrastive Learning for Relation Extraction | https://aclanthology.org/2022.emnlp-main.71/ | [
"William Hogan",
"Jiacheng Li",
"Jingbo Shang"
] | Recent relation extraction (RE) works have shown encouraging improvements by conducting contrastive learning on silver labels generated by distant supervision before fine-tuning on gold labels. Existing methods typically assume all these silver labels are accurate and treat them equally; however, distant supervision is... | 2022.emnlp-main.71 | 10.18653/v1/2022.emnlp-main.71 | null | 2205.12491 | title_snapshot | [
-0.007093590684235096,
-0.0009077577269636095,
0.011880116537213326,
0.06273434311151505,
0.03311501815915108,
0.006388599053025246,
0.0061782896518707275,
0.0021710454020649195,
-0.009503278881311417,
0.017365045845508575,
-0.007503465749323368,
0.050369489938020706,
-0.054560258984565735,
... |
Curriculum Prompt Learning with Self-Training for Abstractive Dialogue Summarization | https://aclanthology.org/2022.emnlp-main.72/ | [
"Changqun Li",
"Linlin Wang",
"Xin Lin",
"Gerard de Melo",
"Liang He"
] | Succinctly summarizing dialogue is a task of growing interest, but inherent challenges, such as insufficient training data and low information density impede our ability to train abstractive models. In this work, we propose a novel curriculum-based prompt learning method with self-training to address these problems. Sp... | 2022.emnlp-main.72 | 10.18653/v1/2022.emnlp-main.72 | null | null | null | [
-0.022472718730568886,
-0.05223735421895981,
-0.019941408187150955,
0.06915966421365738,
0.03786761686205864,
0.007818861864507198,
0.015895823016762733,
0.007808492053300142,
-0.0060636429116129875,
0.011147668585181236,
-0.04988131299614906,
0.043800801038742065,
-0.045435067266225815,
-... |
Zero-Shot Text Classification with Self-Training | https://aclanthology.org/2022.emnlp-main.73/ | [
"Ariel Gera",
"Alon Halfon",
"Eyal Shnarch",
"Yotam Perlitz",
"Liat Ein-Dor",
"Noam Slonim"
] | Recent advances in large pretrained language models have increased attention to zero-shot text classification. In particular, models finetuned on natural language inference datasets have been widely adopted as zero-shot classifiers due to their promising results and off-the-shelf availability. However, the fact that su... | 2022.emnlp-main.73 | 10.18653/v1/2022.emnlp-main.73 | null | 2210.17541 | title_snapshot | [
0.006432629656046629,
-0.03868090361356735,
-0.02478916384279728,
0.065071702003479,
0.04549747332930565,
0.029386376962065697,
0.04222053289413452,
0.01425572857260704,
-0.0029187446925789118,
0.011180994100868702,
-0.030106881633400917,
0.05639716237783432,
-0.06863787025213242,
0.003024... |
Deconfounding Legal Judgment Prediction for European Court of Human Rights Cases Towards Better Alignment with Experts | https://aclanthology.org/2022.emnlp-main.74/ | [
"Santosh T.y.s.s",
"Shanshan Xu",
"Oana Ichim",
"Matthias Grabmair"
] | This work demonstrates that Legal Judgement Prediction systems without expert-informed adjustments can be vulnerable to shallow, distracting surface signals that arise from corpus construction, case distribution, and confounding factors. To mitigate this, we use domain expertise to strategically identify statistically ... | 2022.emnlp-main.74 | 10.18653/v1/2022.emnlp-main.74 | null | 2210.13836 | title_snapshot | [
0.008803632110357285,
-0.024542247876524925,
-0.034068163484334946,
0.026323838159441948,
0.028490271419286728,
-0.008553873747587204,
0.03731561079621315,
-0.018180621787905693,
-0.01739182136952877,
-0.0176447294652462,
-0.03199571371078491,
0.04392268508672714,
-0.08307956904172897,
0.0... |
SQuALITY: Building a Long-Document Summarization Dataset the Hard Way | https://aclanthology.org/2022.emnlp-main.75/ | [
"Alex Wang",
"Richard Yuanzhe Pang",
"Angelica Chen",
"Jason Phang",
"Samuel R. Bowman"
] | Summarization datasets are often assembled either by scraping naturally occurring public-domain summaries—which are nearly always in difficult-to-work-with technical domains—or by using approximate heuristics to extract them from everyday text—which frequently yields unfaithful summaries. In this work, we turn to a slo... | 2022.emnlp-main.75 | 10.18653/v1/2022.emnlp-main.75 | null | 2205.11465 | title_snapshot | [
-0.008427352644503117,
-0.04105570539832115,
-0.023189937695860863,
0.05923360213637352,
0.04141261801123619,
-0.009667564183473587,
0.007484228350222111,
0.013357811607420444,
-0.029482172802090645,
-0.028413575142621994,
-0.03698866441845894,
0.01981276087462902,
-0.05189080163836479,
-0... |
MetaASSIST: Robust Dialogue State Tracking with Meta Learning | https://aclanthology.org/2022.emnlp-main.76/ | [
"Fanghua Ye",
"Xi Wang",
"Jie Huang",
"Shenghui Li",
"Samuel Stern",
"Emine Yilmaz"
] | Existing dialogue datasets contain lots of noise in their state annotations. Such noise can hurt model training and ultimately lead to poor generalization performance. A general framework named ASSIST has recently been proposed to train robust dialogue state tracking (DST) models. It introduces an auxiliary model to ge... | 2022.emnlp-main.76 | 10.18653/v1/2022.emnlp-main.76 | null | 2210.12397 | title_snapshot | [
-0.007059263531118631,
-0.01918714866042137,
-0.009545375593006611,
0.049754876643419266,
0.009872409515082836,
0.04992140829563141,
0.0403905026614666,
-0.00946598406881094,
-0.01661493442952633,
-0.02348872274160385,
-0.03607311099767685,
0.04870198294520378,
-0.07255853712558746,
-0.031... |
Multilingual Machine Translation with Hyper-Adapters | https://aclanthology.org/2022.emnlp-main.77/ | [
"Christos Baziotis",
"Mikel Artetxe",
"James Cross",
"Shruti Bhosale"
] | Multilingual machine translation suffers from negative interference across languages. A common solution is to relax parameter sharing with language-specific modules like adapters. However, adapters of related languages are unable to transfer information, and their total number of parameters becomes prohibitively expens... | 2022.emnlp-main.77 | 10.18653/v1/2022.emnlp-main.77 | null | 2205.10835 | title_snapshot | [
-0.014759331941604614,
-0.006825200282037258,
0.021622728556394577,
0.029809460043907166,
0.018282625824213028,
0.02067233994603157,
0.01880665309727192,
-0.008231033571064472,
-0.015572970733046532,
-0.01952601782977581,
-0.02147143892943859,
0.018305273726582527,
-0.06616180390119553,
0.... |
Z-LaVI: Zero-Shot Language Solver Fueled by Visual Imagination | https://aclanthology.org/2022.emnlp-main.78/ | [
"Yue Yang",
"Wenlin Yao",
"Hongming Zhang",
"Xiaoyang Wang",
"Dong Yu",
"Jianshu Chen"
] | Large-scale pretrained language models have made significant advances in solving downstream language understanding tasks. However, they generally suffer from reporting bias, the phenomenon describing the lack of explicit commonsense knowledge in written text, e.g., ”an orange is orange”. To overcome this limitation, we... | 2022.emnlp-main.78 | 10.18653/v1/2022.emnlp-main.78 | null | 2210.12261 | title_snapshot | [
0.008146927691996098,
-0.000490875041577965,
-0.027302201837301254,
0.03990815579891205,
0.038444407284259796,
0.0457552894949913,
0.03623181954026222,
0.03456224128603935,
-0.027747690677642822,
-0.005025156773626804,
-0.046725083142519,
0.01618018001317978,
-0.08427862823009491,
0.014106... |
Using Commonsense Knowledge to Answer Why-Questions | https://aclanthology.org/2022.emnlp-main.79/ | [
"Yash Kumar Lal",
"Niket Tandon",
"Tanvi Aggarwal",
"Horace Liu",
"Nathanael Chambers",
"Raymond Mooney",
"Niranjan Balasubramanian"
] | Answering questions in narratives about why events happened often requires commonsense knowledge external to the text. What aspects of this knowledge are available in large language models? What aspects can be made accessible via external commonsense resources? We study these questions in the context of answering quest... | 2022.emnlp-main.79 | 10.18653/v1/2022.emnlp-main.79 | null | null | null | [
-0.019765162840485573,
-0.022501084953546524,
-0.026159485802054405,
0.045684412121772766,
0.04013456776738167,
-0.003166917944326997,
0.03346838802099228,
0.019128838554024696,
-0.036182988435029984,
0.028495417907834053,
-0.029980475082993507,
0.05243637040257454,
-0.055285610258579254,
... |
Affective Idiosyncratic Responses to Music | https://aclanthology.org/2022.emnlp-main.80/ | [
"Sky CH-Wang",
"Evan Li",
"Oliver Li",
"Smaranda Muresan",
"Zhou Yu"
] | Affective responses to music are highly personal. Despite consensus that idiosyncratic factors play a key role in regulating how listeners emotionally respond to music, precisely measuring the marginal effects of these variables has proved challenging. To address this gap, we develop computational methods to measure af... | 2022.emnlp-main.80 | 10.18653/v1/2022.emnlp-main.80 | null | 2210.09396 | title_snapshot | [
0.016481736674904823,
-0.00569134159013629,
-0.02113441191613674,
0.0032064805272966623,
0.013643644750118256,
0.015994252637028694,
0.04714690148830414,
0.009900784119963646,
-0.008816882036626339,
-0.023397542536258698,
-0.014337671920657158,
-0.011633225716650486,
-0.050499703735113144,
... |
Successive Prompting for Decomposing Complex Questions | https://aclanthology.org/2022.emnlp-main.81/ | [
"Dheeru Dua",
"Shivanshu Gupta",
"Sameer Singh",
"Matt Gardner"
] | Answering complex questions that require making latent decisions is a challenging task, especially when limited supervision is available. Recent works leverage the capabilities of large language models (LMs) to perform complex question answering in a few-shot setting by demonstrating how to output intermediate rational... | 2022.emnlp-main.81 | 10.18653/v1/2022.emnlp-main.81 | null | 2212.04092 | title_snapshot | [
-0.0013271573698148131,
-0.02623164840042591,
-0.006924073211848736,
0.05743081122636795,
0.04151984676718712,
0.029656661674380302,
0.02397172525525093,
-0.012750119902193546,
-0.015057573094964027,
0.015314140357077122,
-0.019027914851903915,
0.010843728668987751,
-0.06226164475083351,
-... |
Maieutic Prompting: Logically Consistent Reasoning with Recursive Explanations | https://aclanthology.org/2022.emnlp-main.82/ | [
"Jaehun Jung",
"Lianhui Qin",
"Sean Welleck",
"Faeze Brahman",
"Chandra Bhagavatula",
"Ronan Le Bras",
"Yejin Choi"
] | Pre-trained language models (LMs) struggle with consistent reasoning; recently, prompting LMs to generate explanations that self-guide the inference has emerged as a promising direction to amend this. However, these approaches are fundamentally bounded by the correctness of explanations, which themselves are often nois... | 2022.emnlp-main.82 | 10.18653/v1/2022.emnlp-main.82 | null | 2205.11822 | title_snapshot | [
-0.036844827234745026,
-0.023239215835928917,
-0.020068855956196785,
0.03758448362350464,
0.047503989189863205,
0.047525860369205475,
0.040880780667066574,
-0.008750289678573608,
-0.03750871121883392,
0.018974315375089645,
-0.026626214385032654,
0.046950653195381165,
-0.04446223005652428,
... |
DANLI: Deliberative Agent for Following Natural Language Instructions | https://aclanthology.org/2022.emnlp-main.83/ | [
"Yichi Zhang",
"Jianing Yang",
"Jiayi Pan",
"Shane Storks",
"Nikhil Devraj",
"Ziqiao Ma",
"Keunwoo Yu",
"Yuwei Bao",
"Joyce Chai"
] | Recent years have seen an increasing amount of work on embodied AI agents that can perform tasks by following human language instructions. However, most of these agents are reactive, meaning that they simply learn and imitate behaviors encountered in the training data. These reactive agents are insufficient for long-ho... | 2022.emnlp-main.83 | 10.18653/v1/2022.emnlp-main.83 | null | 2210.12485 | title_snapshot | [
-0.012457465752959251,
-0.0312604121863842,
-0.03724142163991928,
0.026815492659807205,
0.011489645577967167,
0.021293701604008675,
0.04764031991362572,
0.014384206384420395,
-0.02424180507659912,
-0.031938716769218445,
-0.04529184475541115,
0.0267063919454813,
-0.06874740868806839,
-0.009... |
Tracing Semantic Variation in Slang | https://aclanthology.org/2022.emnlp-main.84/ | [
"Zhewei Sun",
"Yang Xu"
] | The meaning of a slang term can vary in different communities. However, slang semantic variation is not well understood and under-explored in the natural language processing of slang. One existing view argues that slang semantic variation is driven by culture-dependent communicative needs. An alternative view focuses o... | 2022.emnlp-main.84 | 10.18653/v1/2022.emnlp-main.84 | null | 2210.08635 | title_snapshot | [
-0.015222631394863129,
-0.01404933724552393,
-0.0311418566852808,
0.01440293900668621,
0.028414398431777954,
0.04069030284881592,
0.057417791336774826,
0.04034340754151344,
-0.017129814252257347,
-0.012239331379532814,
-0.0427141897380352,
0.006439120043069124,
-0.046160317957401276,
-0.00... |
Fine-grained Category Discovery under Coarse-grained supervision with Hierarchical Weighted Self-contrastive Learning | https://aclanthology.org/2022.emnlp-main.85/ | [
"Wenbin An",
"Feng Tian",
"Ping Chen",
"Siliang Tang",
"Qinghua Zheng",
"QianYing Wang"
] | Novel category discovery aims at adapting models trained on known categories to novel categories. Previous works only focus on the scenario where known and novel categories are of the same granularity.In this paper, we investigate a new practical scenario called Fine-grained Category Discovery under Coarse-grained supe... | 2022.emnlp-main.85 | 10.18653/v1/2022.emnlp-main.85 | null | 2210.07733 | title_snapshot | [
-0.036106206476688385,
-0.0203808955848217,
0.016984106972813606,
0.04033742472529411,
0.048904940485954285,
0.020429104566574097,
-0.0022330591455101967,
-0.014226611703634262,
-0.035210829228162766,
-0.017678704112768173,
-0.007566533051431179,
0.01126417238265276,
-0.032751210033893585,
... |
PLM-based World Models for Text-based Games | https://aclanthology.org/2022.emnlp-main.86/ | [
"Minsoo Kim",
"Yeonjoon Jung",
"Dohyeon Lee",
"Seung-won Hwang"
] | World models have improved the ability of reinforcement learning agents to operate in a sample efficient manner, by being trained to predict plausible changes in the underlying environment. As the core tasks of world models are future prediction and commonsense understanding, our claim is that pre-trained language mode... | 2022.emnlp-main.86 | 10.18653/v1/2022.emnlp-main.86 | null | null | null | [
-0.03005564957857132,
-0.010166504420340061,
0.0010646810987964272,
0.0325922816991806,
0.03551368787884712,
0.02679174579679966,
0.00891045667231083,
0.021349681541323662,
-0.026504626497626305,
-0.010301172733306885,
-0.034542180597782135,
0.02968583069741726,
-0.057764116674661636,
-0.0... |
Prompt-Based Meta-Learning For Few-shot Text Classification | https://aclanthology.org/2022.emnlp-main.87/ | [
"Haoxing Zhang",
"Xiaofeng Zhang",
"Haibo Huang",
"Lei Yu"
] | Few-shot Text Classification predicts the semantic label of a given text with a handful of supporting instances. Current meta-learning methods have achieved satisfying results in various few-shot situations. Still, they often require a large amount of data to construct many few-shot tasks for meta-training, which is no... | 2022.emnlp-main.87 | 10.18653/v1/2022.emnlp-main.87 | null | null | null | [
-0.01271974015980959,
-0.028493909165263176,
-0.01499489787966013,
0.051949258893728256,
0.028241699561476707,
-0.007222347892820835,
0.03276548534631729,
0.001911517116241157,
-0.03542226180434227,
0.023625316098332405,
-0.03891516104340553,
0.039953719824552536,
-0.06719257682561874,
-0.... |
How well can Text-to-Image Generative Models understand Ethical Natural Language Interventions? | https://aclanthology.org/2022.emnlp-main.88/ | [
"Hritik Bansal",
"Da Yin",
"Masoud Monajatipoor",
"Kai-Wei Chang"
] | Text-to-image generative models have achieved unprecedented success in generating high-quality images based on natural language descriptions. However, it is shown that these models tend to favor specific social groups when prompted with neutral text descriptions (e.g., ‘a photo of a lawyer’). Following Zhao et al. (202... | 2022.emnlp-main.88 | 10.18653/v1/2022.emnlp-main.88 | null | 2210.15230 | title_snapshot | [
0.01838298700749874,
-0.039717089384794235,
-0.04464185610413551,
0.06281667202711105,
0.023369181901216507,
0.013618961907923222,
0.031820155680179596,
0.024009302258491516,
-0.013097099028527737,
-0.03548037260770798,
-0.056445199996232986,
0.03256550431251526,
-0.07556099444627762,
-0.0... |
Geographic Citation Gaps in NLP Research | https://aclanthology.org/2022.emnlp-main.89/ | [
"Mukund Rungta",
"Janvijay Singh",
"Saif M. Mohammad",
"Diyi Yang"
] | In a fair world, people have equitable opportunities to education, to conduct scientific research, to publish, and to get credit for their work, regardless of where they live. However, it is common knowledge among researchers that a vast number of papers accepted at top NLP venues come from a handful of western countri... | 2022.emnlp-main.89 | 10.18653/v1/2022.emnlp-main.89 | null | 2210.14424 | title_snapshot | [
-0.012221446260809898,
-0.033881399780511856,
-0.040097061544656754,
0.016982054337859154,
0.028909869492053986,
-0.0022271503694355488,
0.054682180285453796,
0.04317199066281319,
-0.0346698984503746,
-0.003697848180308938,
-0.0349341481924057,
-0.023450229316949844,
-0.05036507546901703,
... |
Language Models of Code are Few-Shot Commonsense Learners | https://aclanthology.org/2022.emnlp-main.90/ | [
"Aman Madaan",
"Shuyan Zhou",
"Uri Alon",
"Yiming Yang",
"Graham Neubig"
] | We address the general task of structured commonsense reasoning: given a natural language input, the goal is to generate a graph such as an event or a reasoning-graph.To employ large language models (LMs) for this task, existing approaches ‘serialize’ the output graph as a flat list of nodes and edges.Although feasible... | 2022.emnlp-main.90 | 10.18653/v1/2022.emnlp-main.90 | null | 2210.07128 | title_snapshot | [
-0.00013004595530219376,
-0.02850411646068096,
-0.03219187259674072,
0.06528140604496002,
0.04197953641414642,
0.014449778012931347,
0.03196323290467262,
0.024899302050471306,
-0.013750436715781689,
0.012588785029947758,
-0.021343078464269638,
0.04008488729596138,
-0.06241438537836075,
-0.... |
Numerical Optimizations for Weighted Low-rank Estimation on Language Models | https://aclanthology.org/2022.emnlp-main.91/ | [
"Ting Hua",
"Yen-Chang Hsu",
"Felicity Wang",
"Qian Lou",
"Yilin Shen",
"Hongxia Jin"
] | Singular value decomposition (SVD) is one of the most popular compression methods that approximate a target matrix with smaller matrices. However, standard SVD treats the parameters within the matrix with equal importance, which is a simple but unrealistic assumption. The parameters of a trained neural network model ma... | 2022.emnlp-main.91 | 10.18653/v1/2022.emnlp-main.91 | null | 2211.09718 | title_judge | [
-0.06255478411912918,
-0.01254336629062891,
0.011378323659300804,
0.03330378979444504,
0.018706118687987328,
0.08279677480459213,
0.024383487179875374,
-0.02332201600074768,
-0.03794528543949127,
-0.021557632833719254,
-0.02945241704583168,
0.02609761618077755,
-0.05167436599731445,
-0.003... |
Generative Multi-hop Retrieval | https://aclanthology.org/2022.emnlp-main.92/ | [
"Hyunji Lee",
"Sohee Yang",
"Hanseok Oh",
"Minjoon Seo"
] | A common practice for text retrieval is to use an encoder to map the documents and the query to a common vector space and perform a nearest neighbor search (NNS); multi-hop retrieval also often adopts the same paradigm, usually with a modification of iteratively reformulating the query vector so that it can retrieve di... | 2022.emnlp-main.92 | 10.18653/v1/2022.emnlp-main.92 | null | 2204.13596 | title_snapshot | [
-0.01683289371430874,
-0.042617928236722946,
-0.026583202183246613,
0.06717204302549362,
0.05042550340294838,
0.009159781038761139,
0.013945226557552814,
0.010677263140678406,
-0.005708975251764059,
-0.039501339197158813,
-0.014777234755456448,
0.0031596862245351076,
-0.055870916694402695,
... |
Visual Spatial Description: Controlled Spatial-Oriented Image-to-Text Generation | https://aclanthology.org/2022.emnlp-main.93/ | [
"Yu Zhao",
"Jianguo Wei",
"ZhiChao Lin",
"Yueheng Sun",
"Meishan Zhang",
"Min Zhang"
] | Image-to-text tasks such as open-ended image captioning and controllable image description have received extensive attention for decades. Here we advance this line of work further, presenting Visual Spatial Description (VSD), a new perspective for image-to-text toward spatial semantics. Given an image and two objects i... | 2022.emnlp-main.93 | 10.18653/v1/2022.emnlp-main.93 | null | 2210.11109 | title_snapshot | [
0.02263026125729084,
0.0014876023633405566,
0.01217979658395052,
0.05584520846605301,
0.025478119030594826,
0.0018488881178200245,
0.03077642060816288,
0.032894566655159,
-0.007456083316355944,
-0.03502791002392769,
-0.05435434356331825,
-0.0003766945155803114,
-0.04862052947282791,
0.0099... |
M3: A Multi-View Fusion and Multi-Decoding Network for Multi-Document Reading Comprehension | https://aclanthology.org/2022.emnlp-main.94/ | [
"Liang Wen",
"Houfeng Wang",
"Yingwei Luo",
"Xiaolin Wang"
] | Multi-document reading comprehension task requires collecting evidences from different documents for answering questions. Previous research works either use the extractive modeling method to naively integrate the scores from different documents on the encoder side or use the generative modeling method to collect the cl... | 2022.emnlp-main.94 | 10.18653/v1/2022.emnlp-main.94 | null | null | null | [
-0.00038403007783927023,
0.0009136971202678978,
-0.011165601201355457,
0.05207257345318794,
0.05340125039219856,
0.0005535112577490509,
0.010172029957175255,
0.00990370288491249,
-0.02500947006046772,
-0.000933513161726296,
-0.004757297690957785,
0.04923773556947708,
-0.06148383021354675,
... |
COCO-DR: Combating the Distribution Shift in Zero-Shot Dense Retrieval with Contrastive and Distributionally Robust Learning | https://aclanthology.org/2022.emnlp-main.95/ | [
"Yue Yu",
"Chenyan Xiong",
"Si Sun",
"Chao Zhang",
"Arnold Overwijk"
] | We present a new zero-shot dense retrieval (ZeroDR) method, COCO-DR, to improve the generalization ability of dense retrieval by combating the distribution shifts between source training tasks and target scenarios. To mitigate the impact of document differences, COCO-DR continues pretraining the language model on the t... | 2022.emnlp-main.95 | 10.18653/v1/2022.emnlp-main.95 | null | 2210.15212 | title_judge | [
-0.021162046119570732,
-0.024791711941361427,
-0.017276400700211525,
0.08045031875371933,
0.01126807276159525,
0.018288208171725273,
-0.007688137702643871,
0.0013777267886325717,
-0.0004993379698134959,
-0.016105247661471367,
-0.02148967608809471,
0.021491950377821922,
-0.02671159617602825,
... |
Language Model Pre-Training with Sparse Latent Typing | https://aclanthology.org/2022.emnlp-main.96/ | [
"Liliang Ren",
"Zixuan Zhang",
"Han Wang",
"Clare Voss",
"ChengXiang Zhai",
"Heng Ji"
] | Modern large-scale Pre-trained Language Models (PLMs) have achieved tremendous success on a wide range of downstream tasks. However, most of the LM pre-training objectives only focus on text reconstruction, but have not sought to learn latent-level interpretable representations of sentences. In this paper, we manage to... | 2022.emnlp-main.96 | 10.18653/v1/2022.emnlp-main.96 | null | 2210.12582 | title_snapshot | [
-0.00226234900765121,
-0.034887854009866714,
-0.013971802778542042,
0.03184447064995766,
0.046476997435092926,
0.055251896381378174,
0.025860775262117386,
0.02381153404712677,
-0.045949582010507584,
-0.007971490733325481,
-0.007174374535679817,
0.010380866937339306,
-0.05870118737220764,
0... |
On the Transformation of Latent Space in Fine-Tuned NLP Models | https://aclanthology.org/2022.emnlp-main.97/ | [
"Nadir Durrani",
"Hassan Sajjad",
"Fahim Dalvi",
"Firoj Alam"
] | We study the evolution of latent space in fine-tuned NLP models. Different from the commonly used probing-framework, we opt for an unsupervised method to analyze representations. More specifically, we discover latent concepts in the representational space using hierarchical clustering. We then use an alignment function... | 2022.emnlp-main.97 | 10.18653/v1/2022.emnlp-main.97 | null | 2210.12696 | title_snapshot | [
-0.04559549689292908,
-0.02986198477447033,
-0.030408604070544243,
0.040167856961488724,
0.05583197623491287,
0.021448802202939987,
0.03969617187976837,
0.026255415752530098,
-0.012325489893555641,
-0.01236797496676445,
-0.01972980983555317,
0.013814247213304043,
-0.04491497576236725,
-0.0... |
Watch the Neighbors: A Unified K-Nearest Neighbor Contrastive Learning Framework for OOD Intent Discovery | https://aclanthology.org/2022.emnlp-main.98/ | [
"Yutao Mou",
"Keqing He",
"Pei Wang",
"Yanan Wu",
"Jingang Wang",
"Wei Wu",
"Weiran Xu"
] | Discovering out-of-domain (OOD) intent is important for developing new skills in task-oriented dialogue systems. The key challenges lie in how to transfer prior in-domain (IND) knowledge to OOD clustering, as well as jointly learn OOD representations and cluster assignments. Previous methods suffer from in-domain overf... | 2022.emnlp-main.98 | 10.18653/v1/2022.emnlp-main.98 | null | 2210.08909 | title_snapshot | [
-0.018499374389648438,
-0.02627500146627426,
0.010123108513653278,
0.03594154864549637,
0.013589197769761086,
0.01794186234474182,
0.0271486546844244,
-0.00011683234333759174,
0.0067801885306835175,
0.01079986896365881,
-0.024213427677750587,
0.0336402989923954,
-0.06514620035886765,
-0.02... |
Extracted BERT Model Leaks More Information than You Think! | https://aclanthology.org/2022.emnlp-main.99/ | [
"Xuanli He",
"Lingjuan Lyu",
"Chen Chen",
"Qiongkai Xu"
] | The collection and availability of big data, combined with advances in pre-trained models (e.g. BERT), have revolutionized the predictive performance of natural language processing tasks. This allows corporations to provide machine learning as a service (MLaaS) by encapsulating fine-tuned BERT-based models as APIs. Due... | 2022.emnlp-main.99 | 10.18653/v1/2022.emnlp-main.99 | null | 2210.11735 | title_snapshot | [
-0.015084565617144108,
0.007973209023475647,
-0.013200177811086178,
0.060379382222890854,
0.036418378353118896,
-0.007357910741120577,
0.04476218298077583,
-0.017726177349686623,
-0.001598292263224721,
0.013682601042091846,
-0.02630329690873623,
0.03595143184065819,
-0.046099793165922165,
... |
Do Vision-and-Language Transformers Learn Grounded Predicate-Noun Dependencies? | https://aclanthology.org/2022.emnlp-main.100/ | [
"Mitja Nikolaus",
"Emmanuelle Salin",
"Stephane Ayache",
"Abdellah Fourtassi",
"Benoit Favre"
] | Recent advances in vision-and-language modeling have seen the development of Transformer architectures that achieve remarkable performance on multimodal reasoning tasks.Yet, the exact capabilities of these black-box models are still poorly understood. While much of previous work has focused on studying their ability to... | 2022.emnlp-main.100 | 10.18653/v1/2022.emnlp-main.100 | null | 2210.12079 | title_snapshot | [
-0.011271734721958637,
0.010468622669577599,
-0.003705727867782116,
0.0502997562289238,
0.02691948413848877,
0.014234514907002449,
0.01950729638338089,
0.030903929844498634,
-0.022931799292564392,
0.003782673506066203,
-0.05059272050857544,
0.06395985186100006,
-0.050140123814344406,
-0.01... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.