Behavior Gated Language Models

Panayiotis Georgiou; Prashanth Gurunath Shivakumar; Shao-Yen Tseng; Shrikanth Narayanan

REVIEW 3 major objections 6 minor 28 references

Behavior Gated Language Models

T0 review · 3 major / 6 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read This paper claims that language models predict words better when a fixed behavior classifier's outputs gate the LM's hidden state, and reports lower perplexity on three corpora.

desk verdict Small but consistent LM gains from behavior gating; the paper needs to state explicitly whether the behavior model is causal before the results can be trusted. read the letter →

arxiv 1909.00107 v1 pith:PKZ34QCJ submitted 2019-08-31 cs.CL

Prashanth Gurunath Shivakumar , Shao-Yen Tseng , Panayiotis Georgiou , Shrikanth Narayanan This is my paper · ORCID

classification cs.CL

keywords behavior-gatedlanguagemodelpsycholinguisticcontextLSTMperplexitymulti-labelbehaviorclassificationgatingmechanismcouplestherapycorpusPennTreebank

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

The paper proposes a way to give an LSTM language model a second channel of information: the speaker's inferred behavioral state. A pre-trained multi-label classifier recognizes five behaviors—acceptance, blame, negativity, positivity, sadness—from the text, and its time-synced outputs are used to gate the language model's hidden state before the next-word prediction. On the Couples Therapy Corpus, the Cancer Couples Interaction Dataset, and Penn Treebank, the behavior-gated models report lower test perplexity than their ungated baselines, including an improvement on the previous state-of-the-art AWD-LSTM on Penn Treebank. If the result holds, behavior signals are a cheap, architecture-agnostic way to carry long-range psychological context into word prediction.

What carries the argument

The behavior gate is the load-bearing mechanism: a fixed, pre-trained multi-label RNN behavior classifier (single LSTM layer, dimension 50) that outputs $b_t$ for the five behaviors; a small trainable time-synced RNN turns those abstract behavior outputs into a gating vector $z_t$; and $z_t$ multiplicatively gates the LSTM language model's hidden state before the output projection. The first recurrent layer of the behavior model is initialized from the behavior classifier and frozen, so the added information carries no extra supervision signal of its own during LM training. The gate is what converts "who is speaking and how" into a different word-prediction distribution without changing the underlying LM architecture.

What would settle it

Force the behavior classifier to compute its gate at time $t$ from $x_1$ through $x_{t-1}$ only and rerun the Penn Treebank experiments; if test perplexity returns to 57.3 for the AWD-LSTM baseline, the claimed gains came from future-word leakage.

Watch

Extended reading notes

Core claim

The central claim is that conditioning next-word prediction on behavioral context through a gate improves language modeling. The proposed model writes the prediction as $P(y \mid x, z)$ with $z = f(P(B \mid x))$; the behavior model predicts the joint probability of behavior occurrences from the input text, and the gate combines its output with the LM's hidden state. Reported test perplexities drop from 66.32 to 64.71 on CoupTher, from 159.65 to 148.78 on Cancer, from 82.7 to 78.75 (LSTM-Medium) and 78.4 to 75.80 (LSTM-Large) on PTB, and from 57.3 to 56.92 for AWD-LSTM on PTB. The motivating idea is that psychological states shape vocabulary choices, so a model that knows the state should assign higher probability to the words that state tends to produce.

Load-bearing premise

The load-bearing premise is that the behavior gate at time $t$ is computed only from the words produced up to that point; the paper does not state this, and if the gate sees future words, every reported perplexity gain could be an artifact of leakage.

Editorial extensions

If this is right

Because the behavior classifier is frozen after pretraining, behavior gating adds only a small trainable RNN and can be bolted onto any recurrent LM architecture.
The 6.81% relative improvement on the Cancer Couples dataset, where both models were trained on CoupTher, suggests the behavior signal transfers to behavior-related domains it was not trained on.
The PTB results, including the drop from 57.3 to 56.92 for AWD-LSTM, indicate the benefit is not confined to behavior-rich text and may generalize to ordinary written language.
LSTM-Medium with behavior gating roughly matches the LSTM-Large baseline's test perplexity with far fewer parameters, suggesting behavioral context can substitute for some model capacity.
The paper's planned application to automatic speech recognition would let behavior states derived from a first pass rescore a second pass, effectively using long-range speaker state as a rescoring feature.

Reading between the lines

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

The same gating template could be tested with topic, emotion, or speaker-role classifiers; the paper only reports the five behavior labels, so the general claim about context-gated conditioning is an inference from its results.
A direct stress test would be to shuffle behavior labels across utterances; if perplexity gains persist, the gate is exploiting low-level textual correlates of the labels rather than the behaviors themselves.
For streaming use, a strictly causal version of the gate may lose some of the reported gains, and the paper's own future-work section hints that future behavior states could help in rescoring; quantifying that causal gap would sharpen the practical value.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

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

I read the paper. It is what it looks like: a feature-augmentation study that adds a pre-trained multi-label behavior classifier's outputs as gates to standard RNNLMs. That specific combination is new relative to the POS/topic/conversation-type work they cite. The empirical story is consistent: on CoupTher, Cancer, and PTB with three architectures, behavior gating lowers perplexity. The out-of-domain Cancer result (6.81% relative improvement) is the most convincing piece, since the behavior model was trained on therapy transcripts and still helped on a different, similarly behavior-rich corpus. Credit where due: they used a fixed pretrained behavior encoder, which avoids training the gate from the LM's own target, so circularity is not a real problem.

The soft spots are real but manageable. The largest one is that they never explicitly say the behavior model is causal. Section 2.2 defines it as predicting P(B|x) over the input sequence, which could mean a sequence-level or bidirectional model. If b_t at time t contains information from future words, every perplexity gain in Tables 1 and 2 could be leakage, not behavior modeling. I think the intended architecture is causal—Figure 2 shows time-synced b_t, and the future-work paragraph says they could derive "both past and future behavior states" for ASR rescoring, implying the current model uses only past context. But that inference should be stated in the model definition. A referee should ask for this explicitly.

Second, the PTB gains are tiny—0.66% test perplexity for AWD-LSTM, from 57.3 to 56.92, with no error bars or multiple seeds. That is within plausible run-to-run noise. The abstract's phrase "improve on previous state-of-the-art models" is also stronger than what they actually show: they beat their chosen baselines, but they do not compare against the current PTB SOTA (e.g., Transformer-XL). That claim should be toned down.

Third, no code or data is released, and the gating architecture is only sketched. For a paper whose entire contribution is a new conditioning mechanism, that limits reproducibility.

Overall: if the causality point is resolved and the PTB claim is softened, this is a solid, modest contribution for people working on clinical or dialogue language modeling. It is not a breakthrough, but it is plausible and worth a serious referee. I would not block publication on the small PTB gain alone; the Cancer result and the consistent direction of effect across three datasets carry the paper.

Referee Report

3 major / 6 minor

Summary. The paper proposes augmenting recurrent neural network language models with a fixed, pre-trained behavior classifier whose outputs gate the LM's prediction layer. The behavior model predicts five multi-label behavioral codes (Acceptance, Blame, Negativity, Positivity, Sadness) from text, and the gated LM is evaluated on the Couples Therapy Corpus, the Cancer Couples Interaction Dataset, and the Penn Treebank, using LSTM-Medium, LSTM-Large, and AWD-LSTM architectures. The reported test perplexity improvements range from 0.66% (PTB, AWD-LSTM) to 6.81% (Cancer), and the paper claims that behavior gating consistently lowers perplexity across all settings.

Significance. If the empirical claim is established, the paper offers a novel and potentially useful way to inject human-annotated behavioral state into language models, with plausible applications in clinical and behavioral domains. The work also has strengths: it reports parameter counts, evaluates across multiple architectures and corpora, and fixes the behavior model during LM training, which avoids end-to-end overfitting to the behavior labels. However, the central claim currently rests on an unspecified causal structure for the behavior model and on small performance differences that are presented without control conditions or variance estimates, so the contribution is not yet fully established.

major comments (3)

[§2.2, §2.3.2, Figure 2] The behavior model is defined in §2.2 as predicting P(B|x) over the input sequence, and §2.3.2 defines the gated model as P(y|x,z) with z≡f(P(B|x)); nowhere does the paper state that the behavior output b_t used at time t depends only on x_1,...,x_t or x_1,...,x_{t-1}. If the behavior RNN is bidirectional or otherwise processes the full utterance, then during next-word prediction the gate contains future words, and every perplexity reduction in Tables 1 and 2 could be an artifact of future-context leakage rather than evidence about behavior modeling. Figure 2 and the ASR rescoring remark in Section 5 suggest a causal implementation was intended, but this is not stated or verified anywhere. Please explicitly define the input horizon of the behavior model at each time step and, if the model is causal, state this in §2.3.2 and confirm it in the architecture description.
[§3.3, §4, Table 2] The improvement is attributed specifically to behavioral information, but the gated model also adds a trainable time-synced RNN and receives a 5-dimensional behavior posterior. The gains are small in several settings, especially the PTB AWD-LSTM test improvement of 0.66% (57.3 to 56.92). Without a control condition that uses a randomly initialized behavior network, constant or scrambled gate values, or a behavior model whose outputs are otherwise shown to be informative on the target corpus, it is not clear whether the gains come from the behavior signal itself or from the added trainable capacity and initialization. Please include such a control or otherwise demonstrate that the behavior outputs are the source of the improvement.
[§4.1.1, §4.2, Tables 1 and 2] All reported results appear to be single runs with no standard deviation, number of seeds, or significance test. The smallest claimed improvement, AWD-LSTM on PTB test, is well within the range one would expect from random seed variation for this architecture. Given that the central empirical claim is the presence of consistent perplexity reductions, please report multiple runs with variance or justify that the observed differences are not noise, particularly for the PTB results.

minor comments (6)

[§4.2.1] The sentence 'we employ the AWD-LSTM proposed by Merity et al. (2018) with QRNN instead of LSTM' is confusing because AWD-LSTM is an LSTM-based architecture; please clarify whether the experiments use the AWD-LSTM with LSTM cells or the AWD-QRNN variant.
[Table 2] The parameter counts for '+ Behavior gating' are reported as 20M for LSTM-Medium (same as baseline) while other rows increase by 1M or 3M; please clarify whether these counts include the fixed behavior model and the trainable gate RNN, and how the counts are computed.
[§4.1.1] The phrase 'relative improvement of (1.42% on valid) 0.66% in perplexity' is garbled; it should be rewritten as, for example, 'a relative improvement of 1.42% on validation and 0.66% on test.'
[§3.2] The behavior model section does not specify the input segmentation (utterance-level, fixed window, or full conversation), the multi-label loss function, or how class imbalance among the five behaviors is handled; a few sentences on these details would improve reproducibility.
[Abstract and §4.2] The claim of improving on 'previous state-of-the-art models' should be phrased more cautiously, since the comparison is against one strong baseline (AWD-LSTM) rather than the full current PTB leaderboard.
[§3.1.1] The phrase '10,000 unique entries' is ambiguous; please state whether this is the vocabulary size or the number of utterances/turns, and give the corresponding split sizes for the Cancer dataset.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the behavior gate is a frozen feature extractor trained on human-annotated labels, and the language-model target is not used to train it.

full rationale

The derivation chain is self-contained. The behavior model is pre-trained on human-annotated behavior labels from the Couples Therapy Corpus and then fixed during language-model training (Section 2.3.2: "The first recurrent layer of the behavior model is initialized with pretrained weights from the model described in Section 2.2 and fixed during language modeling training"). The language model still optimizes next-word prediction, and the behavior labels are not derived from the next-word target, so the reported perplexity improvements are not equivalent to a fitted parameter or to the model's own inputs. The claimed improvements are empirical comparisons against vanilla LSTM baselines on held-out test sets, including an out-of-domain evaluation on the Cancer dataset and on PTB, where the behavior model was trained on a different corpus. Citations to the authors' prior work (e.g., Narayanan and Georgiou 2013, Tseng et al. 2016) are used as background motivation and architecture precedent, not as a load-bearing uniqueness theorem or as the sole justification for the central claim. The reader's concern that the behavior model may be non-causal and leak future context is a validity or leakage risk, not a circularity of derivation, because the paper's equations do not define the behavior gate in terms of the prediction target. Thus no circular step meeting the required evidence standard is present.

Assumptions & free parameters 3 free parameters · 3 assumptions · 0 invented entities

The central claim rests on the psycholinguistic hypothesis that behavior states measurably alter word choice, and on the implicit architectural assumption that the behavior model is causal. The paper does not prove either; the first is imported from prior literature, and the second is never articulated.

free parameters (3)

Behavior label set = Acceptance, Blame, Negativity, Positivity, Sadness
Five behaviors selected from the 33 codes in CoupTher (Section 3.2). The selection is justified only by 'good separability' in prior work, so the central result depends on this hand-picked subset.
Behavior model hidden size = 50
Single-layer LSTM of dimension 50 for the pre-trained behavior model (Section 3.2). No sensitivity analysis is provided, so the effect of this choice on the gate quality is unknown.
LM hyperparameters (learning rate, dropout, batch size, behavior model layers) = Not reported; chosen by grid search
Section 3.3 says a grid search was run and the best validation result is reported. This makes the reported perplexity gains selected on the validation set, which can inflate test improvements.

assumptions (3)

domain assumption Different psychological behavior states incite measurable differences in language use.
Motivates the whole approach, imported from Pennebaker and Graybeal (2001) and Lindquist et al. (2015). The paper does not test this premise.
ad hoc to paper The behavior model is causal: behavior outputs at time t use only words up to time t.
Section 2.2 defines the behavior model as predicting P(B|x) over the input sequence, and the paper does not restrict it to past context. The LM comparison is only fair under this causal assumption, which is never stated.
domain assumption Behavior annotations from CoupTher transfer to other corpora, including generic text.
The Cancer and PTB experiments assume a classifier trained on therapy dialogues yields useful behavior signals out of domain. The Cancer result is consistent with this, but the PTB result is small.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Behavior Gated Language Models." pith.science (2026). https://pith.science/paper/PKZ34QCJ

@misc{pith2026190900107,
  author       = {Pith},
  title        = {Pith review of: Behavior Gated Language Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/PKZ34QCJ}},
  note         = {Machine review of arXiv:1909.00107}
}

read the original abstract

Most current language modeling techniques only exploit co-occurrence, semantic and syntactic information from the sequence of words. However, a range of information such as the state of the speaker and dynamics of the interaction might be useful. In this work we derive motivation from psycholinguistics and propose the addition of behavioral information into the context of language modeling. We propose the augmentation of language models with an additional module which analyzes the behavioral state of the current context. This behavioral information is used to gate the outputs of the language model before the final word prediction output. We show that the addition of behavioral context in language models achieves lower perplexities on behavior-rich datasets. We also confirm the validity of the proposed models on a variety of model architectures and improve on previous state-of-the-art models with generic domain Penn Treebank Corpus.

Figures

Figures reproduced from arXiv: 1909.00107 by the authors.

**Figure 1.** RNN language model. tendencies can provide useful information in statistical language modeling. And although not directly related, behavioral information may also correlate with conversation-type and topic. Thus, we propose the use of psycholinguistic behavior signals as a gating mechanism to augment typical language models. Effectively inferring behaviors from sources like spoken text, written articles can lead t… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

28 extracted references · 21 canonical work pages

[1]

URL: " 'urlintro :=

ENTRY address author booktitle chapter edition editor howpublished institution journal key month note number organization pages publisher school series title type volume year eprint doi pubmed url lastchecked label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block STRINGS urlintro eprinturl eprintpr...
[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]

Zico Kolter, and Vladlen Koltun

Shaojie Bai, J. Zico Kolter, and Vladlen Koltun. 2019. Trellis networks for sequence modeling. In International Conference on Learning Representations ( ICLR )

work page 2019
[4]

James Bradbury, Stephen Merity, Caiming Xiong, and Richard Socher. 2017. https://openreview.net/forum?id=H1zJ-v5xl Quasi-recurrent neural networks . In 5th International Conference on Learning Representations, ICLR 2017, Toulon, France, April 24-26, 2017, Conference Track Proceedings

work page 2017
[5]

Christensen, D.C

A. Christensen, D.C. Atkins, S. Berns, J. Wheeler, D.H. Baucom, and L.E. Simpson. 2004. Traditional versus integrative behavioral couple therapy for significantly and chronically distressed married couples . Journal of Consulting and Clinical Psychology, 72

work page 2004
[7]

Dieng, Chong Wang, Jianfeng Gao, and John W

Adji B. Dieng, Chong Wang, Jianfeng Gao, and John W. Paisley. 2017. Topic RNN : A recurrent neural network with long-range semantic dependency. In International Conference on Learning Representations ( ICLR )

work page 2017
[8]

Shalini Ghosh, Oriol Vinyals, Brian Strope, Scott Roy, Tom Dean, and Larry Heck. 2016. Contextual lstm (clstm) models for large scale nlp tasks. arXiv preprint arXiv:1602.06291

arXiv 2016
[9]

James Gibson, Dogan Can, Panayiotis Georgiou, David Atkins, and Shrikanth Narayanan. 2017. Attention networks for modeling behavior in addiction counseling. In Proceedings of Interspeech

work page 2017

Show all 28 references

[10]

Gong , X

C. Gong , X. Li , and X. Wu . 2014. Recurrent neural network language model with part-of-speech for mandarin speech recognition. In The 9th International Symposium on Chinese Spoken Language Processing

2014
[11]

Ben Krause, Emmanuel Kahembwe, Iain Murray, and Steve Renals. 2019. Dynamic evaluation of transformer language models. arXiv preprint arXiv:1904.08378

2019 arXiv
[12]

Kristen A Lindquist, Jennifer K MacCormack, and Holly Shablack. 2015. The role of language in emotion: Predictions from psychological constructionism. Frontiers in Psychology, 6

2015
[13]

Mitchell Marcus, Grace Kim, Mary Ann Marcinkiewicz, Robert MacIntyre, Ann Bies, Mark Ferguson, Karen Katz, and Britta Schasberger. 1994. The penn treebank: annotating predicate argument structure. In Proceedings of the workshop on Human Language Technology. Association for Com...

1994
[14]

Stephen Merity, Nitish Shirish Keskar, and Richard Socher. 2018. Regularizing and optimizing lstm language models. In International Conference on Learning Representations ( ICLR )

2018
[15]

Tom \'a s Mikolov, Anoop Deoras, Stefan Kombrink, Luk \'a s Burget, and Jan C ernock \`y . 2011. Empirical evaluation and combination of advanced language modeling techniques. In Twelfth Annual Conference of the International Speech Communication Association

2011
[16]

Tom \'a s Mikolov, Martin Karafi \'a t, Luk \'a s Burget, Jan C ernock \`y , and Sanjeev Khudanpur. 2010. Recurrent neural network based language model. In Eleventh Annual Conference of the International Speech Communication Association

2010
[17]

Tomas Mikolov and Geoffrey Zweig. 2012. Context dependent recurrent neural network language model. In 2012 IEEE Spoken Language Technology Workshop (SLT), pages 234--239. IEEE

2012
[18]

Narayanan and P

S. Narayanan and P. G. Georgiou. 2013. https://doi.org/10.1109/JPROC.2012.2236291 Behavioral signal processing: Deriving human behavioral informatics from speech and language . Proceedings of the IEEE, PP(99)

2013
[19]

James W Pennebaker and Anna Graybeal. 2001. Patterns of natural language use: Disclosure, personality, and social integration. Current Directions in Psychological Science, 10(3)

2001
[20]

Ofir Press. 2019. Partially shuffling the training data to improve language models. arXiv preprint arXiv:1903.04167

2019 arXiv
[21]

Sutton, Susan Vadaparampil, Richard E

Maija Reblin, Steven K. Sutton, Susan Vadaparampil, Richard E. Heyman, and Lee Ellington. 2018. Behind closed doors: How advanced cancer couples communicate at home. Journal of Psychosocial Oncology, 37

2018
[22]

Yangyang Shi, Pascal Wiggers, and Catholijn M Jonker. 2010. Language modelling with dynamic bayesian networks using conversation types and part of speech information. In The 22nd Benelux Conference on Artificial Intelligence, BNAIC

2010
[23]

Yangyang Shi, Pascal Wiggers, and Catholijn M Jonker. 2012. Towards recurrent neural networks language models with linguistic and contextual features. In Thirteenth Annual Conference of the International Speech Communication Association

2012
[24]

Chao Su, Heyan Huang, Shumin Shi, Yuhang Guo, and Hao Wu. 2017. A parallel recurrent neural network for language modeling with pos tags. In Proceedings of the 31st Pacific Asia Conference on Language, Information and Computation, pages 140--147

2017
[25]

Shao-Yen Tseng, Sandeep Nallan Chakravarthula, Brian Baucom, and Panayiotis Georgiou. 2016. Couples behavior modeling and annotation using low-resource LSTM language models. In Proceedings of Interspeech

2016
[26]

Wenlin Wang, Zhe Gan, Wenqi Wang, Dinghan Shen, Jiaji Huang, Wei Ping, Sanjeev Satheesh, and Lawrence Carin. 2018. Topic compositional neural language model. In International Conference on Artificial Intelligence and Statistics

2018
[27]

Imel, David C

Bo Xiao, James Gibson, Dogan Can, Zac E. Imel, David C. Atkins, Panayiotis Georgiou, and Shrikanth S. Narayanan. 2016. https://doi.org/10.21437/Interspeech.2016-1560 Behavioral coding of therapist language in addiction counseling using recurrent neural networks . In Proceeding...

2016 doi
[28]

Wojciech Zaremba, Ilya Sutskever, and Oriol Vinyals. 2014. Recurrent neural network regularization. arXiv preprint arXiv:1409.2329

2014 arXiv
[29]

ER jTԿ ÛߴoWXs `˃g ' F3 vu l ｵ:OZ

11em plus .33em minus .07em 4000 4000 100 4000 4000 500 `\.=1000 = #1 \@IEEEnotcompsoconly \@IEEEcompsoconly #1 * [1] 0pt [0pt][0pt] #1 * [1] 0pt [0pt][0pt] #1 * \| ** #1 \@IEEEauthorblockNstyle \@IEEEcompsocnotconfonly \@IEEEauthorblockAstyle \@IEEEcompsocnotconfonly \@IEEEco...

2010 arXiv

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [1]

URL: " 'urlintro :=

ENTRY address author booktitle chapter edition editor howpublished institution journal key month note number organization pages publisher school series title type volume year eprint doi pubmed url lastchecked label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block STRINGS urlintro eprinturl eprintpr...

[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]

Zico Kolter, and Vladlen Koltun

Shaojie Bai, J. Zico Kolter, and Vladlen Koltun. 2019. Trellis networks for sequence modeling. In International Conference on Learning Representations ( ICLR )

work page 2019

[4] [4]

James Bradbury, Stephen Merity, Caiming Xiong, and Richard Socher. 2017. https://openreview.net/forum?id=H1zJ-v5xl Quasi-recurrent neural networks . In 5th International Conference on Learning Representations, ICLR 2017, Toulon, France, April 24-26, 2017, Conference Track Proceedings

work page 2017

[5] [5]

Christensen, D.C

A. Christensen, D.C. Atkins, S. Berns, J. Wheeler, D.H. Baucom, and L.E. Simpson. 2004. Traditional versus integrative behavioral couple therapy for significantly and chronically distressed married couples . Journal of Consulting and Clinical Psychology, 72

work page 2004

[6] [7]

Dieng, Chong Wang, Jianfeng Gao, and John W

Adji B. Dieng, Chong Wang, Jianfeng Gao, and John W. Paisley. 2017. Topic RNN : A recurrent neural network with long-range semantic dependency. In International Conference on Learning Representations ( ICLR )

work page 2017

[7] [8]

Shalini Ghosh, Oriol Vinyals, Brian Strope, Scott Roy, Tom Dean, and Larry Heck. 2016. Contextual lstm (clstm) models for large scale nlp tasks. arXiv preprint arXiv:1602.06291

arXiv 2016

[8] [9]

James Gibson, Dogan Can, Panayiotis Georgiou, David Atkins, and Shrikanth Narayanan. 2017. Attention networks for modeling behavior in addiction counseling. In Proceedings of Interspeech

work page 2017