Boosting Short Text Classification with Multi-Source Information Exploration and Dual-Level Contrastive Learning

Fausto Giunchiglia; Lan Huang; Mengyu Li; Renchu Guan; Wei Pang; Xiaoyue Feng; Yonghao Liu

REVIEW 4 major objections 4 minor 62 references

Boosting Short Text Classification with Multi-Source Information Exploration and Dual-Level Contrastive Learning

T0 review · 4 major / 4 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read MI-DELIGHT boosts short-text classification with multi-source graphs and hierarchical dual-level contrastive learning, surpassing prior models and even LLMs on several benchmarks.

desk verdict A solid short-text classification method with credible gains over non-LLM baselines, but the LLM comparison is under-specified and reproducibility details are missing. read the letter →

arxiv 2501.09214 v1 pith:5AW5OPB5 submitted 2025-01-16 cs.CL

Yonghao Liu , Mengyu Li , Wei Pang , Fausto Giunchiglia , Lan Huang , Xiaoyue Feng , Renchu Guan This is my paper · ORCID

classification cs.CL

keywords shorttextclassificationcontrastivelearninggraphneuralnetworksmulti-sourceinformationhierarchicalmulti-tasksemi-supervisedknowledgegraphspseudo-labeling

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Short texts are hard to classify because they are semantically sparse and labeled examples are scarce. MI-DELIGHT attacks both problems at once: it enriches each text with statistical, linguistic, and factual information encoded in three graphs, then uses a hierarchical pair of contrastive learning tasks to pull meaningful neighbors together before classification. The paper reports that this design significantly surpasses previous competitive models and even outperforms popular large language models on three of the five benchmarks, with the largest gains on Snippets, Ohsumed, and TagMyNews. The claim matters because it suggests a comparatively small, label-efficient graph-based model can match or beat much larger pretrained models on domain-specific short text.

What carries the argument

The carrying mechanism is the hierarchical pipeline from fine-grained to coarse-grained contrastive signals. Instance-level contrastive learning (ICL) treats an original text and its WordNet-synonym augmented version as a positive pair and all other texts as negatives, running directly on normalized text embeddings without a projection head. Cluster-level contrastive learning (CCL) then assigns pseudo-labels by connected-component labeling over symmetric nearest-neighbor links in the ICL feature space and applies a supervised contrastive loss with swapped supervision between original and augmented corpora; a projection head maps features into a lower-dimensional space for this coarser task. The final linear classifier consumes the CCL-refined abstract features, so each stage inherits the previous stage's features and adds its own abstraction, making the architecture a hierarchy rather than parallel task combination.

What would settle it

Take the labeled split and compute CCL pseudo-labels from the trained model without using true labels; measure their agreement with ground truth. If agreement is near chance while final accuracy stays high, the reported gains are not explained by semantic clustering. Alternatively, replace CCL pseudo-labels with random but fixed cluster assignments and rerun; if accuracy remains similar to the full model, the CCL contribution is not due to cluster semantics.

Watch

Extended reading notes

Core claim

On its own terms, the paper establishes that combining multi-source graph-based text enrichment with a dual-level contrastive learning hierarchy is a superior architecture for semi-supervised short text classification. Specifically, MI-DELIGHT builds a word graph (PMI-weighted GloVe), a POS graph, and an entity graph (TransE embeddings for NELL-linked entities), aggregates text representations by TF-IDF-weighted pooling, and then runs instance-level contrastive learning followed by cluster-level contrastive learning with pseudo-labels from symmetric nearest-neighbor connected components, finally classifying from the abstract features. Across Twitter, MR, Snippets, Ohsumed, and TagMyNews with only tens of labeled examples per class, the paper reports accuracy and macro-F1 improvements over all four baseline families, with the clearest margins on the three datasets where unlabeled data is relatively abundant. The ablation shows every component—each graph, each contrastive level, and the hierarchical ordering—contributes to the final result.

Load-bearing premise

The cluster-level contrastive step assumes that texts connected through symmetric nearest-neighbor links in the current embedding space form semantically meaningful clusters; if those pseudo-labels are mostly wrong, the auxiliary task can propagate noise into the classifier.

Editorial extensions

If this is right

On Snippets, Ohsumed, and TagMyNews, MI-DELIGHT outperforms GPT-3.5, Bloom-7.1B, Llama2-7B, and Llama3-8B in both accuracy and macro-F1.
Removing any single graph (word, POS, or entity) hurts accuracy; removing the word graph causes the largest drop.
Dual-level contrastive learning beats each level alone, and the hierarchical ordering beats a parallel arrangement of the same tasks.
WordNet synonym augmentation is preferable to random deletion or BERT-context substitution for generating positive pairs.
The method's advantage grows with the amount of unlabeled data, consistent with self-supervised signal extraction being the main driver.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

Because the paper does not report confidence filtering on CCL pseudo-labels, a natural extension would be to weight cluster-level loss by cluster compactness or to blend soft pseudo-labels, which might make the gains robust to noisy nearest-neighbor links.
The staged ICL-to-CCL-to-classification design could transfer to other semi-supervised graph tasks beyond text, such as node classification or few-shot relation extraction, wherever instance- and cluster-level structure can be ordered by complexity.
A direct test of whether CCL is truly learning semantics would be to compare pseudo-labels against ground-truth labels on the unlabeled split: agreement well above chance would support the mechanism, while near-chance agreement with high downstream accuracy would suggest the benefit comes from regularization rather than clustering.
The LLM comparison fine-tunes only about 7B parameter models due to compute limits, so the short-text advantage could narrow as larger or fully tuned LLMs are used.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

MI-DELIGHT is a solid empirical contribution to low-resource short text classification. The headline that it beats all non-LLM baselines on five standard benchmarks is credible; the headline that it beats LLMs is not verifiable from what is written. The hierarchical ordering of instance-level, cluster-level, then classification is a modest but real architectural idea, and the ablations back up each design choice.

What is genuinely new: the combination of three graph sources (word, POS, entity) with dual-level contrastive learning in a causal chain. The paper does well by running a wide set of baselines, including the strongest recent GNN-contrastive models, and showing consistent gains. The ablation set is thorough—removing any graph or contrastive level hurts, and the hierarchical structure beats the parallel version. Code is provided. Ten runs with pairwise t-tests claimed, which is good practice.

Soft spots, in order of importance. First, the LLM comparison is under-specified. The abstract says MI-DELIGHT 'even outperforms popular large language models on several datasets,' but the only protocol detail is 'we only fine-tune approximately 7B LLMs through some GPU reduction techniques.' GPT-3.5 is not 7B, so we don't know if it was fine-tuned, zero-shot, or few-shot. No LoRA rank, epochs, learning rate, prompt template, or training example count for Bloom, Llama2, Llama3. And the table caption excludes LLMs from the bold significance test. So the LLM claim is not established. That is a real weakness, but it does not undermine the non-LLM comparison, which is the paper's core. Second, no standard deviations or hyperparameter values (tau, eta, zeta, GCN layers) are reported despite claims of t-tests. This hurts reproducibility more than the claims. Third, the classification head equation is ambiguous: 'linear transformation followed by ReLU' before cross-entropy, with Q dimension equal to the number of classes, suggests a missing softmax. Minor, but should be cleaned up. Fourth, the CCL pseudo-labels from nearest-neighbor components could be noisy, and no confidence filtering is applied. The ablation shows CCL helps, so this is a concern, not a fatal flaw.

One note: the strongest baseline, GIFT, is actually a prior paper from the same group (Liu et al. 2024b), not independent work. That's fine—self-comparison is expected—but it means the state-of-the-art claim is really about beating their own previous method by a small margin plus older methods.

Who this is for: researchers in short text classification, GNN-based text representation, or contrastive learning. It deserves a serious referee. The method is sound enough, the experiments are mostly supportive, and the issues are addressable with more detail. I would send it out.

Referee Report

4 major / 4 minor

Summary. The paper proposes MI-DELIGHT, a short text classification model that combines three types of graph-based multi-source information (word/POS/entity graphs), instance- and cluster-level contrastive learning, and a hierarchical task architecture. On five benchmark datasets with limited labeled samples, the model outperforms a wide range of published baselines, and the ablations indicate that each component and the hierarchical ordering contribute to performance. The abstract also claims that MI-DELIGHT outperforms popular large language models on several datasets.

Significance. If the empirical claims hold, MI-DELIGHT would be a strong new state of the art for low-label short text classification, with particularly large gains on Snippets, Ohsumed, and TagMyNews. The paper's strengths are its broad baseline coverage, including the recent GIFT method that MI-DELIGHT exceeds, ablations that separately validate the graphs, both contrastive levels, and the hierarchical design, and the public code link. However, the central claim of superiority over LLMs is not supported by a verifiable protocol, and the reported significance testing is not backed by variance statistics. The pseudo-label noise concern for cluster-level contrastive learning is real but is at least partially mitigated by the ablation showing that removing CCL hurts accuracy; it does not by itself invalidate the empirical comparison.

major comments (4)

[Experiment, Baselines; Abstract] The claim that MI-DELIGHT 'even outperforms popular large language models on several datasets' is not supported by a reproducible protocol. The only description is that 'approximately 7B LLMs' were fine-tuned 'through some GPU reduction techniques,' which is ambiguous for Llama2/Llama3/Bloom and inapplicable to GPT-3.5, whose size and fine-tuning status are not stated. No prompt template, number of training examples, epochs, learning rate, or parameter-efficient method (e.g., LoRA rank) are given. Since Table 2 explicitly excludes LLMs from the bold 'best performance' significance test, the LLM comparison is not statistically analyzed. The authors should either specify the complete LLM protocol or remove the LLM claim from the abstract and conclusions.
[Table 2 caption; Experiment, Evaluation Metric] The paper states that the best non-LLM performance is highlighted 'based on the pairwise t-test with 95% confidence,' and that all experiments were repeated ten times, but no standard deviations, p-values, or confidence intervals are reported anywhere. A t-test result is not checkable from the table as presented. Please report per-run standard deviations (or error bars) and, if the claim is to be maintained, the corresponding significance values or at least a statement of how the pairwise comparison was conducted.
[Method, Eq. (6)] The cluster-level contrastive loss in Eq. (6) appears to include the anchor itself as a positive example. The outer sum over j includes j = i, and the indicator I_{Y_ij}=1 is satisfied for j = i because an instance is in the same connected component as itself; the denominator explicitly excludes k = i. This makes the numerator contain exp(U_i · U_i / tau) = exp(1/tau), which is likely unintended. Please add an explicit exclusion of j = i or clarify the intended indexing for the positive set.
[Experiment, Model Performance and Model Variants; Reproducibility] The experimental setup is missing key hyperparameters and training details: the temperatures tau for ICL and CCL, the loss weights eta and zeta in Eq. (8), GCN hidden dimensions and number of layers, projection head dimensions, the WordNet synonym replacement ratio, and optimization settings (epochs, batch size, learning rate, scheduler). Without these values the reported numbers cannot be reproduced from the description alone, even with the code link. In addition, Table 3 reports Ohsumed F1 = 32.20 for MI-DELIGHT (WordNet) while Table 2 reports Ohsumed F1 = 33.20 for the same full model; this inconsistency should be resolved.

minor comments (4)

[Method, Classification Task, Eq. (7)] Equation (7) applies cross-entropy directly to Q, but the text says Q is obtained by a linear transformation followed by a ReLU activation. Cross-entropy normally operates on probabilities, so the description likely omits a softmax over Q; please clarify whether Q contains logits or normalized probabilities.
[Method, Hierarchical Structure among Tasks] The phrase 'casual relationships among tasks' should be 'causal relationships' (the same typo appears in the Introduction).
[Experiment, Baselines] It is unclear whether the BERT-avg and BERT-cls baselines are fine-tuned on the training data or used as fixed feature extractors; this affects the fairness of the comparison. Please specify the protocol for these baselines.
[Method, Cluster-Level Contrastive Learning] The symbol Y is used both for the pseudo-label matrix in CCL and for the ground-truth one-hot labels in Eq. (7); using distinct symbols (e.g., Y^pseudo and Y^label) would prevent confusion.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: empirical gains are measured against external benchmarks and independent baselines.

full rationale

MI-DELIGHT is an empirical systems paper: the central claim is that a particular architecture with three constructed graphs, ICL and CCL auxiliary losses, and a hierarchical task order reaches higher accuracy/macro-F1 than published baselines on five external benchmark datasets. No component is defined in terms of the held-out labels or the test metric. The CCL pseudo-labels are generated by connected components on symmetric nearest-neighbor links in the current text embedding space (Eq. 5-6), which is a self-supervised bootstrap, not an injection of the target variable; the classification loss (Eq. 7) uses only the 40-per-class training labels. The comparison to GIFT and other baselines is against independent prior work, and the paper's own ablations show each module contributes positively rather than being forced by construction. The only notable weakness is the under-specified LLM baseline protocol ('Due to computational resource constraints, we only fine-tune approximately 7B LLMs through some GPU reduction techniques'), which makes the 'outperforms popular large language models' claim difficult to verify; however, that is a reproducibility/experimental-design issue, not a circularity. No equation reduces to its own input and no load-bearing claim rests on a self-citation chain, so the circularity score is 0.

Assumptions & free parameters 5 free parameters · 6 assumptions · 0 invented entities

MI-DELIGHT introduces no new physical or mathematical entities. Its contribution is architectural: combining known graph and contrastive techniques. However, the method relies on several domain assumptions (pseudo-label quality, augmentation fidelity, entity linking accuracy) and on multiple unstated hyperparameters, which are the real 'free parameters' of the paper.

free parameters (5)

Temperature tau for ICL and CCL
Appears in Eq. 3 and Eq. 6 for both contrastive losses; value is not reported in the paper.
Loss weights eta and zeta
Weights in the final loss Eq. 8; no values or sensitivity analysis are given.
GCN hidden dimensions and number of layers = 2 layers
The paper specifies a 2-layer GCN but not the hidden dimensionality, which materially affects learned representations.
Projection head dimension = half of Z dimension
Described as half the dimension of Z for the CCL projection head, but the base dimension and training details are unspecified.
WordNet synonym augmentation ratio
The number or fraction of words replaced by synonyms during data augmentation is not specified.

assumptions (6)

standard math GCN message passing (Kipf and Welling 2017) produces useful node representations for text graphs.
Used as the backbone for all three graphs via Eq. 1; treated as a proven standard operation.
domain assumption PMI over co-occurrence and TF-IDF aggregation capture useful statistical and linguistic signals in short texts.
The word and POS graph edges rely on PMI and the text aggregation relies on TF-IDF; the paper does not validate this choice independently.
domain assumption WordNet synonym substitution preserves the semantic label of a short text.
Data augmentation in the Method section assumes that replacing words with WordNet synonyms keeps the text's meaning and class unchanged; this is not verified per dataset.
domain assumption TAGME entity linking against the NELL knowledge graph returns correct and useful entities for short texts.
The factual information graph depends on this; the paper does not report linking quality and TAGME is typically designed for Wikipedia, not NELL.
ad hoc to paper Connected components over symmetric nearest-neighbor links in embedding space form meaningful pseudo-clusters.
The CCL loss uses these pseudo-labels; the validity of this clustering is not measured and is central to the proposed method.
ad hoc to paper Task complexity and feature abstraction increase from ICL to CCL to classification, so hierarchical ordering is beneficial.
The paper asserts this causal relationship in the Introduction and Method, and supports it only with a single ablation against a parallel variant.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Boosting Short Text Classification with Multi-Source Information Exploration and Dual-Level Contrastive Learning." pith.science (2026). https://pith.science/paper/5AW5OPB5

@misc{pith2026250109214,
  author       = {Pith},
  title        = {Pith review of: Boosting Short Text Classification with Multi-Source Information Exploration and Dual-Level Contrastive Learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5AW5OPB5}},
  note         = {Machine review of arXiv:2501.09214}
}

read the original abstract

Short text classification, as a research subtopic in natural language processing, is more challenging due to its semantic sparsity and insufficient labeled samples in practical scenarios. We propose a novel model named MI-DELIGHT for short text classification in this work. Specifically, it first performs multi-source information (i.e., statistical information, linguistic information, and factual information) exploration to alleviate the sparsity issues. Then, the graph learning approach is adopted to learn the representation of short texts, which are presented in graph forms. Moreover, we introduce a dual-level (i.e., instance-level and cluster-level) contrastive learning auxiliary task to effectively capture different-grained contrastive information within massive unlabeled data. Meanwhile, previous models merely perform the main task and auxiliary tasks in parallel, without considering the relationship among tasks. Therefore, we introduce a hierarchical architecture to explicitly model the correlations between tasks. We conduct extensive experiments across various benchmark datasets, demonstrating that MI-DELIGHT significantly surpasses previous competitive models. It even outperforms popular large language models on several datasets.

Figures

Figures reproduced from arXiv: 2501.09214 by the authors.

**Figure 1.** The overall architecture of MI-DELIGHT. We first generate augmented samples for the input texts. Then, the original [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

62 extracted references · 55 canonical work pages

[1]

, " * write output.state after.block = add.period write newline

ENTRY address archivePrefix author booktitle chapter edition editor eid eprint howpublished institution isbn journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION init.state.consts #0 'before.a...
[2]

write newline

" write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...
[3]

AI@Meta. 2024. Llama 3 Model Card

2024
[4]

Bordes, A.; Usunier, N.; Garcia-Duran, A.; Weston, J.; and Yakhnenko, O. 2013. Translating embeddings for modeling multi-relational data. In NeurIPS

work page 2013
[5]

Carlson, A.; Betteridge, J.; Kisiel, B.; Settles, B.; Jr., E. R. H.; and Mitchell, T. M. 2010. Toward an Architecture for Never-Ending Language Learning. In Fox, M.; and Poole, D., eds., AAAI

work page 2010
[6]

Caron, M.; Misra, I.; Mairal, J.; Goyal, P.; Bojanowski, P.; and Joulin, A. 2020. Unsupervised learning of visual features by contrasting cluster assignments. In NeurIPS

work page 2020
[7]

Chang, Y.; Wang, X.; Wang, J.; Wu, Y.; Zhu, K.; Chen, H.; Yang, L.; Yi, X.; Wang, C.; Wang, Y.; et al. 2023. A survey on evaluation of large language models. arXiv preprint arXiv:2307.03109

arXiv 2023
[8]

Chen, J.; Hu, Y.; Liu, J.; Xiao, Y.; and Jiang, H. 2019. Deep short text classification with knowledge powered attention. In AAAI

work page 2019

Show all 62 references

[9]

Chen, J.; Zhang, R.; Mao, Y.; and Xu, J. 2022. Contrastnet: A contrastive learning framework for few-shot text classification. In AAAI

2022
[10]

Devlin, J.; Chang, M.-W.; Lee, K.; and Toutanova, K. 2019. BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. In NAACL

2019
[11]

Di Stefano, L.; and Bulgarelli, A. 1999. A simple and efficient connected components labeling algorithm. In ICIAP

1999
[12]

Dilrukshi, I.; De Zoysa, K.; and Caldera, A. 2013. Twitter news classification using SVM. In ICCSE

2013
[13]

Ding, K.; Wang, J.; Li, J.; Li, D.; and Liu, H. 2020. Be More with Less: Hypergraph Attention Networks for Inductive Text Classification. In EMNLP

2020
[14]

Edunov, S.; Ott, M.; Auli, M.; and Grangier, D. 2018. Understanding Back-Translation at Scale. In EMNLP

2018
[15]

Gao, T.; Yao, X.; and Chen, D. 2021. SimCSE: Simple Contrastive Learning of Sentence Embeddings. In EMNLP

2021
[16]

Guan, R.; Liu, Y.; Feng, X.; and Li, X. 2021. VPALG: Paper-publication Prediction with Graph Neural Networks. In CIKM

2021
[17]

Hersh, W.; Buckley, C.; Leone, T.; and Hickam, D. 1994. OHSUMED: An interactive retrieval evaluation and new large test collection for research. In SIGIR

1994
[18]

Hu, L.; Yang, T.; Shi, C.; Ji, H.; and Li, X. 2019. Heterogeneous graph attention networks for semi-supervised short text classification. In EMNLP-IJNLP

2019
[19]

Huang, L.; Ma, D.; Li, S.; Zhang, X.; and Wang, H. 2019. Text Level Graph Neural Network for Text Classification. In EMNLP-IJNLP

2019
[20]

R.; Maire, M.; and Khademi, M

Huynh, T.; Kornblith, S.; Walter, M. R.; Maire, M.; and Khademi, M. 2022. Boosting contrastive self-supervised learning with false negative cancellation. In CVPR

2022
[21]

Kenter, T.; and De Rijke, M. 2015. Short text similarity with word embeddings. In CIKM

2015
[22]

Khosla, P.; Teterwak, P.; Wang, C.; Sarna, A.; Tian, Y.; Isola, P.; Maschinot, A.; Liu, C.; and Krishnan, D. 2020. Supervised contrastive learning. In NeurIPS

2020
[23]

Kim, Y. 2014. Convolutional Neural Networks for Sentence Classification. In EMNLP

2014
[24]

N.; and Welling, M

Kipf, T. N.; and Welling, M. 2017. Semi-supervised classification with graph convolutional networks. In ICLR

2017
[25]

Li, M.; Liu, Y.; Giunchiglia, F.; Feng, X.; and Guan, R. 2024. Simple-Sampling and Hard-Mixup with Prototypes to Rebalance Contrastive Learning for Text Classification. arxiv preprint arXiv:2405.11524

2024
[26]

S.; and He, L

Li, Q.; Peng, H.; Li, J.; Xia, C.; Yang, R.; Sun, L.; Yu, P. S.; and He, L. 2022. A Survey on Text Classification: From Traditional to Deep Learning. ACM TIST, 13(2): 1--41

2022
[27]

F.; Gardner, M.; Belinkov, Y.; Peters, M

Liu, N. F.; Gardner, M.; Belinkov, Y.; Peters, M. E.; and Smith, N. A. 2019. Linguistic Knowledge and Transferability of Contextual Representations. In NAACL

2019
[28]

Liu, P.; Qiu, X.; Chen, X.; Wu, S.; and Huang, X.-J. 2015. Multi-timescale long short-term memory neural network for modelling sentences and documents. In EMNLP

2015
[29]

Liu, P.; Qiu, X.; and Huang, X. 2016. Recurrent neural network for text classification with multi-task learning. In IJCAI

2016
[30]

Liu, P.; Qiu, X.; and Huang, X.-J. 2017. Adversarial Multi-task Learning for Text Classification. In ACL

2017
[31]

Liu, X.; You, X.; Zhang, X.; Wu, J.; and Lv, P. 2020. Tensor graph convolutional networks for text classification. In AAAI

2020
[32]

Liu, Y.; Guan, R.; Giunchiglia, F.; Liang, Y.; and Feng, X. 2021. Deep attention diffusion graph neural networks for text classification. In EMNLP

2021
[33]

Liu, Y.; Huang, L.; Cao, B.; Li, X.; Giunchiglia, F.; Feng, X.; and Guan, R. 2024 a . A simple but effective approach for unsupervised few-shot graph classification. In WWW

2024
[34]

Liu, Y.; Huang, L.; Giunchiglia, F.; Feng, X.; and Guan, R. 2024 b . Improved Graph Contrastive Learning for Short Text Classification. In AAAI

2024
[35]

Liu, Y.; Li, M.; Li, X.; Giunchiglia, F.; Feng, X.; and Guan, R. 2022. Few-shot node classification on attributed networks with graph meta-learning. In SIGIR

2022
[36]

Liu, Y.; Li, M.; Li, X.; Guan, R.; and Feng, X. 2023 a . Local and Global: Temporal Question Answering via Information Fusion. In IJCAI

2023
[37]

Liu, Y.; Li, M.; Li, X.; Huang, L.; Giunchiglia, F.; Liang, Y.; Feng, X.; and Guan, R. 2024 c . Meta-GPS++: Enhancing Graph Meta-Learning with Contrastive Learning and Self-Training. ACM TKDD, 18(9): 1--30

2024
[38]

Liu, Y.; Li, M.; Liang, D.; Li, X.; Giunchiglia, F.; Huang, L.; Feng, X.; and Guan, R. 2024 d . Resolving Word Vagueness with Scenario-guided Adapter for Natural Language Inference. In IJCAI

2024
[39]

Liu, Y.; Liang, D.; Fang, F.; Wang, S.; Wu, W.; and Jiang, R. 2023 b . Time-aware multiway adaptive fusion network for temporal knowledge graph question answering. In ICASSP

2023
[40]

Ouyang, L.; Wu, J.; Jiang, X.; Almeida, D.; Wainwright, C.; Mishkin, P.; Zhang, C.; Agarwal, S.; Slama, K.; Ray, A.; et al. 2022. Training language models to follow instructions with human feedback. In NeurIPS

2022
[41]

Pan, L.; Hang, C.-W.; Sil, A.; and Potdar, S. 2022. Improved text classification via contrastive adversarial training. In AAAI

2022
[42]

Pang, B.; and Lee, L. 2005. Seeing Stars: Exploiting Class Relationships for Sentiment Categorization with Respect to Rating Scales. In ACL

2005
[43]

Phan, X.-H.; Nguyen, L.-M.; and Horiguchi, S. 2008. Learning to classify short and sparse text & web with hidden topics from large-scale data collections. In WWW

2008
[44]

L.; Fan, A.; Akiki, C.; Pavlick, E.; Ili \'c , S.; Hesslow, D.; Castagn \'e , R.; Luccioni, A

Scao, T. L.; Fan, A.; Akiki, C.; Pavlick, E.; Ili \'c , S.; Hesslow, D.; Castagn \'e , R.; Luccioni, A. S.; Yvon, F.; et al. 2022. Bloom: A 176b-parameter open-access multilingual language model. arXiv preprint arXiv:2211.05100

2022 arXiv
[45]

I.; Yu, J.; Shi, L.; and Al Moubayed, N

Su, Z.; Harit, A.; Cristea, A. I.; Yu, J.; Shi, L.; and Al Moubayed, N. 2022. Contrastive learning with heterogeneous graph attention networks on short text classification. In IJCNN

2022
[46]

Tang, J.; Qu, M.; and Mei, Q. 2015. Pte: Predictive text embedding through large-scale heterogeneous text networks. In SIGKDD

2015
[47]

Thilakaratne, M.; Falkner, K.; and Atapattu, T. 2019. A systematic review on literature-based discovery: general overview, methodology, & statistical analysis. CSUR, 52(6): 1--34

2019
[48]

Tian, Y.; Krishnan, D.; and Isola, P. 2020. Contrastive multiview coding. In ECCV

2020
[49]

Touvron, H.; Martin, L.; Stone, K.; Albert, P.; Almahairi, A.; Babaei, Y.; Bashlykov, N.; Batra, S.; Bhargava, P.; Bhosale, S.; et al. 2023. Llama 2: Open foundation and fine-tuned chat models. arXiv preprint arXiv:2307.09288

2023 arXiv
[50]

Vitale, D.; Ferragina, P.; and Scaiella, U. 2012. Classification of short texts by deploying topical annotations. In ECIR

2012
[51]

Wang, J.; Wang, Z.; Zhang, D.; and Yan, J. 2017. Combining Knowledge with Deep Convolutional Neural Networks for Short Text Classification. In IJCAI

2017
[52]

Wang, Y.; Wang, S.; Yao, Q.; and Dou, D. 2021. Hierarchical Heterogeneous Graph Representation Learning for Short Text Classification. In EMNLP

2021
[53]

Wei, J.; and Zou, K. 2019. EDA: Easy Data Augmentation Techniques for Boosting Performance on Text Classification Tasks. In EMNLP-IJNLP

2019
[54]

Wu, X.; Gao, C.; Zang, L.; Han, J.; Wang, Z.; and Hu, S. 2022. Esimcse: Enhanced sample building method for contrastive learning of unsupervised sentence embedding. In COLING

2022
[55]

I.; Li, J.; L \'e vy, D.; Nie, A.; Jurafsky, D.; and Ng, A

Xie, Z.; Wang, S. I.; Li, J.; L \'e vy, D.; Nie, A.; Jurafsky, D.; and Ng, A. Y. 2017. Data noising as smoothing in neural network language models. In ICLR

2017
[56]

Yang, T.; Hu, L.; Shi, C.; Ji, H.; Li, X.; and Nie, L. 2021. HGAT: Heterogeneous graph attention networks for semi-supervised short text classification. ACM TOIS, 39(3): 1--29

2021
[57]

Yao, L.; Mao, C.; and Luo, Y. 2019. Graph convolutional networks for text classification. In AAAI

2019
[58]

Ye, Z.; Jiang, G.; Liu, Y.; Li, Z.; and Yuan, J. 2020. Document and word representations generated by graph convolutional network and bert for short text classification. In ECAI

2020
[59]

R.; and King, I

Zeng, J.; Li, J.; Song, Y.; Gao, C.; Lyu, M. R.; and King, I. 2018. Topic Memory Networks for Short Text Classification. In EMNLP

2018
[60]

Zhang, Y.; and Yang, Q. 2021. A survey on multi-task learning. IEEE TKDE, 34(12): 5586--5609

2021
[61]

Zhang, Y.; Yu, X.; Cui, Z.; Wu, S.; Wen, Z.; and Wang, L. 2020. Every Document Owns Its Structure: Inductive Text Classification via Graph Neural Networks. In ACL

2020
[62]

Zheng, M.; Wang, F.; You, S.; Qian, C.; Zhang, C.; Wang, X.; and Xu, C. 2021. Weakly supervised contrastive learning. In CVPR

2021

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.

[1] [1]

, " * write output.state after.block = add.period write newline

ENTRY address archivePrefix author booktitle chapter edition editor eid eprint howpublished institution isbn journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION init.state.consts #0 'before.a...

[2] [2]

write newline

" write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...

[3] [3]

AI@Meta. 2024. Llama 3 Model Card

2024

[4] [4]

Bordes, A.; Usunier, N.; Garcia-Duran, A.; Weston, J.; and Yakhnenko, O. 2013. Translating embeddings for modeling multi-relational data. In NeurIPS

work page 2013

[5] [5]

Carlson, A.; Betteridge, J.; Kisiel, B.; Settles, B.; Jr., E. R. H.; and Mitchell, T. M. 2010. Toward an Architecture for Never-Ending Language Learning. In Fox, M.; and Poole, D., eds., AAAI

work page 2010

[6] [6]

Caron, M.; Misra, I.; Mairal, J.; Goyal, P.; Bojanowski, P.; and Joulin, A. 2020. Unsupervised learning of visual features by contrasting cluster assignments. In NeurIPS

work page 2020

[7] [7]

Chang, Y.; Wang, X.; Wang, J.; Wu, Y.; Zhu, K.; Chen, H.; Yang, L.; Yi, X.; Wang, C.; Wang, Y.; et al. 2023. A survey on evaluation of large language models. arXiv preprint arXiv:2307.03109

arXiv 2023

[8] [8]

Chen, J.; Hu, Y.; Liu, J.; Xiao, Y.; and Jiang, H. 2019. Deep short text classification with knowledge powered attention. In AAAI

work page 2019