Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Detailed Action
Claims 1-19 are pending. Claims 1, 9, and 12 are independent.
Claims 2-8 depend from Claim 1.
Claims 10-11 depend from Claim 9.
Claims 13-19 depend from Claim 12.
This Application was published as U.S. 2026/0162671.
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 5 Dec 2024 are in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1-3 and 9-14 are rejected under 35 U.S.C. 103 as being unpatentable in view of Khorshid et al. (US2024/0119932 hereinafter Khorshid) in view of Xu et al. (Xiao Xu, Libo Qin, Kaiji Chen, Guoxing Wu, Linlin Li, and Wanxiang Che, “Text Is No More Enough! A Benchmark for Profile-Based Spoken Language Understanding,” in AAAI, 2022 hereinafter Xu)
With regards to claim 1, Khorshid teaches:
generating feature vectors for word tokens in an input utterance by incorporating profile information into an utterance encoder; [Khorshid Fig 3 teaches assistant system where natural-language understanding (NLU) module (210) extract’s information from a “user profile” … [and] “formulating the ranked n-grams into features” …[to] “identify one or more of a domain, an intent, or a slot from the user input in a personalized and context-aware manner” (Par [0090]) where “n-gram may be represented in a d-dimensional vector space” (Par [0207]; see Fig 19) and thus the features formulated are vectors. While Khorshid does not specifically state an encoder, and encoder is often included as part of an NLU module (See Xu Fig 3) The inputs are passed to the “entity resolution module 212 may tokenize text by text normalization, extract syntax features from text, and extract semantic features from text based on NLP” (Par [0097]) which generates word tokens for the n-grams in an input utterance.]
With regards to claim 1, Khorshid fails to teach:
A method for profile-based intent detection and slot filling using a joint model for profile-based intent detection and slot filling with slot-to-intent attention (a JPIS model), comprising:
utilizing the generated feature vectors with a slot-to-intent attention to derive label-specific vector representations of intents and slot labels;
calculating a similarity feature matrix between intent and slot labels via label-specific vector representations;
utilizing the similarity feature matrix for computing an attention weight vector;
computing a weighted sum vector based on the attention weight vector and intent label-specific vectors;
predicting an intent label of the input utterance by taking the weighted sum vector as input into an intent decoder; and
predicting a slot label for each word token in the input utterance using a slot decoder that leverages the representation of the predicted intent label and the generated feature vectors from the utterance encoder.
With regards to claim 1, Xu teaches:
A method for profile-based intent detection and slot filling using a joint model for profile-based intent detection and slot filling with slot-to-intent attention (a JPIS model), comprising: [Xu Fig 3 teaches Profile based SLU model using a joint model for profile based intent detection using intent decoder, and performing slot filling by using slot decoder with attention based encoder to intent decoder.]
utilizing the generated feature vectors with a slot-to-intent attention to derive label-specific vector representations of intents and slot labels; calculating a similarity feature matrix between intent and slot labels via label-specific vector representations; [Xu Fig 3 teaches self-attention mechanism connected intent to slot and deriving label-specific vector representations and calculating similarity feature matrix, which is taught by Equation 6 (page 5), and where mathematical combination of q, W, and h in Equation 6 meets the claimed limitations]
utilizing the similarity feature matrix for computing an attention weight vector; [Xu teaches Equation 6, page 5]
computing a weighted sum vector based on the attention weight vector and intent label-specific vectors; [Xu teaches Equation 7 (page 5)]
predicting an intent label of the input utterance by taking the weighted sum vector as input into an intent decoder; and [Xu teaches Equation 14 (page 5 and 8)]
predicting a slot label for each word token in the input utterance using a slot decoder that leverages the representation of the predicted intent label and the generated feature vectors from the utterance encoder. [Xu teaches Equation 18 (pages 5 and 8)]
It would be obvious to one of ordinary skill at the time of Applicant’s filing to combine the assistant system using user profile as taught by Khorshid with the profile based spoken language understanding system using joint intent and slot filling as taught by Xu. The motivation to combine the teachings of Khorshid with Xu is because Xu teaches “a multi-level knowledge adapter to equip the existing SLU models with the ability to incorporate profile information, which has the following advantages: (1) it achieves a fine-grained knowledge injection for both sentence-level intent detection and token-level slot filling; (2) it can be used as a plugin and
easily be compatible with the existing state-of-the-art SLU models” (Sec 1, page 2) which increases the capabilities of Khorshid to provide better profile information in its system.]
With regards to claim 2, Khorshid in view of Xu teaches:
All the limitations of claim 1
further comprising employing slot label-specific vectors in the slot-to-intent attention to guide intent detection, resulting in a weighted sum vector of intent label-specific vectors. [Xu teaches Equation 7 (page 5)]
With regards to claim 3, Khorshid in view of Xu teaches:
All the limitations of claim 2
wherein using the weight sum vector as input to the intent decoder. [Xu teaches multi-knowledge adapter as a plugin using weighted sum vector as a input for the intent decoder (Sec 4.3 Page 5)
With regards to claim 9, Khorshid teaches:
A profile-based intent detection and slot filling system, comprising: an utterance encoder incorporating profile information to generate feature vectors for word tokens in an input utterance; [Khorshid Fig 3 teaches assistant system where natural-language understanding (NLU) module (210) extract’s information from a “user profile” … [and] “formulating the ranked n-grams into features” …[to] “identify one or more of a domain, an intent, or a slot from the user input in a personalized and context-aware manner” (Par [0090]) where “n-gram may be represented in a d-dimensional vector space” (Par [0207]; see Fig 19) and thus the features formulated are vectors. While Khorshid does not specifically state an encoder, and encoder is often included as part of an NLU module (See Xu Fig 3) The inputs are passed to the “entity resolution module 212 may tokenize text by text normalization, extract syntax features from text, and extract semantic features from text based on NLP” (Par [0097]) which generates word tokens for the n-grams in an input utterance.]
With regards to claim 9, Khorshid fails to teach:
a slot-to-intent attention utilizing the generated feature vectors to derive label-specific vector representations of intents and slot labels;
utilizing the label-specific vector representations to calculate a similarity feature matrix between intent and slot labels;
computing an attention weight vector based on the similarity feature matrix;
utilizing the attention weight vector and intent label-specific vectors for computing a weighted sum vector;
an intent decoder taking the weighted sum vector as input to predict an intent label of the input utterance; and
a slot decoder leveraging the representation of the predicted intent label and the generated feature vectors from the utterance encoder to predict a slot label for each word token in the input utterance.
With regards to claim 9, Xu teaches:
a slot-to-intent attention utilizing the generated feature vectors to derive label-specific vector representations of intents and slot labels; utilizing the label-specific vector representations to calculate a similarity feature matrix between intent and slot labels; [Xu Fig 3 teaches self-attention mechanism connected intent to slot and deriving label-specific vector representations and calculating similarity feature matrix, which is taught by Equation 6 (page 5), and where mathematical combination of q, W, and h in Equation 6 meets the claimed limitations]
computing an attention weight vector based on the similarity feature matrix; [Xu teaches Equation 6, page 5]
utilizing the attention weight vector and intent label-specific vectors for computing a weighted sum vector; [Xu teaches Equation 7 (page 5)]
an intent decoder taking the weighted sum vector as input to predict an intent label of the input utterance; and [Xu teaches Equation 14 (page 5 and 8)]
a slot decoder leveraging the representation of the predicted intent label and the generated feature vectors from the utterance encoder to predict a slot label for each word token in the input utterance. [Xu teaches Equation 18 (pages 5 and 8)]
It would be obvious to one of ordinary skill at the time of Applicant’s filing to combine the assistant system using user profile as taught by Khorshid with the profile based spoken language understanding system using joint intent and slot filling as taught by Xu. The motivation to combine the teachings of Khorshid with Xu is because Xu teaches “a multi-level knowledge adapter to equip the existing SLU models with the ability to incorporate profile information, which has the following advantages: (1) it achieves a fine-grained knowledge injection for both sentence-level intent detection and token-level slot filling; (2) it can be used as a plugin and
easily be compatible with the existing state-of-the-art SLU models” (Sec 1, page 2) which increases the capabilities of Khorshid to provide better profile information in its system.]
Claim 10 is a system claim with limitations corresponding to the limitations of method Claim 2 and is rejected under similar rationale.
Claim 11 is a system claim with limitations corresponding to the limitations of method Claim 3 and is rejected under similar rationale.
With regards to claim 12, Khorshid teaches:
A non-transitory computer readable storage medium tangibly embodying a computer readable program code having computer readable instructions that, when executed, causes a computer device to carry out [Khorshid teaches computer-readable non-transitory storage medium (Par [0237])]
generating feature vectors for word tokens in an input utterance by incorporating profile information into an utterance encoder; [Khorshid Fig 3 teaches assistant system where natural-language understanding (NLU) module (210) extract’s information from a “user profile” … [and] “formulating the ranked n-grams into features” …[to] “identify one or more of a domain, an intent, or a slot from the user input in a personalized and context-aware manner” (Par [0090]) where “n-gram may be represented in a d-dimensional vector space” (Par [0207]; see Fig 19) and thus the features formulated are vectors. While Khorshid does not specifically state an encoder, and encoder is often included as part of an NLU module (See Xu Fig 3) The inputs are passed to the “entity resolution module 212 may tokenize text by text normalization, extract syntax features from text, and extract semantic features from text based on NLP” (Par [0097]) which generates word tokens for the n-grams in an input utterance.]
With regards to claim 12, Khorshid fails to teach:
a method for profile-based intent detection and slot filling using a joint model for profile-based intent detection and slot filling with slot-to-intent attention (a JPIS model), the method comprising:
utilizing the generated feature vectors with a slot-to-intent attention to derive label-specific vector representations of intents and slot labels; calculating a similarity feature matrix between intent and slot labels via label-specific vector representations;
utilizing the similarity feature matrix for computing an attention weight vector;
computing a weighted sum vector based on the attention weight vector and intent label-specific vectors;
predicting an intent label of the input utterance by taking the weighted sum vector as input into an intent decoder; and
predicting a slot label for each word token in the input utterance using a slot decoder that leverages the representation of the predicted intent label and the generated feature vectors from the utterance encoder.
With regards to claim 12, Xu teaches:
a method for profile-based intent detection and slot filling using a joint model for profile-based intent detection and slot filling with slot-to-intent attention (a JPIS model), the method comprising: [Xu Fig 3 teaches Profile based SLU model using a joint model for profile based intent detection using intent decoder, and performing slot filling by using slot decoder with attention based encoder to intent decoder.]
utilizing the generated feature vectors with a slot-to-intent attention to derive label-specific vector representations of intents and slot labels; calculating a similarity feature matrix between intent and slot labels via label-specific vector representations;
[Xu Fig 3 teaches self-attention mechanism connected intent to slot and deriving label-specific vector representations and calculating similarity feature matrix, which is taught by Equation 6 (page 5), and where mathematical combination of q, W, and h in Equation 6 meets the claimed limitations]
utilizing the similarity feature matrix for computing an attention weight vector; [Xu teaches Equation 6, page 5]
computing a weighted sum vector based on the attention weight vector and intent label-specific vectors; [Xu teaches Equation 7 (page 5)]
predicting an intent label of the input utterance by taking the weighted sum vector as input into an intent decoder; and [Xu teaches Equation 14 (page 5 and 8)]
predicting a slot label for each word token in the input utterance using a slot decoder that leverages the representation of the predicted intent label and the generated feature vectors from the utterance encoder. [Xu teaches Equation 18 (pages 5 and 8)]
It would be obvious to one of ordinary skill at the time of Applicant’s filing to combine the assistant system using user profile as taught by Khorshid with the profile based spoken language understanding system using joint intent and slot filling as taught by Xu. The motivation to combine the teachings of Khorshid with Xu is because Xu teaches “a multi-level knowledge adapter to equip the existing SLU models with the ability to incorporate profile information, which has the following advantages: (1) it achieves a fine-grained knowledge injection for both sentence-level intent detection and token-level slot filling; (2) it can be used as a plugin and
easily be compatible with the existing state-of-the-art SLU models” (Sec 1, page 2) which increases the capabilities of Khorshid to provide better profile information in its system.]
Claim 13 is a non-transitory computer readable storage medium claim with limitations corresponding to the limitations of method Claim 2 and is rejected under similar rationale.
Claim 14 is a non-transitory computer readable storage medium claim with limitations corresponding to the limitations of method Claim 3 and is rejected under similar rationale.
Allowable Subject Matter
Claims 4-8 and 15-19 objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Joseph J Yamamoto whose telephone number is (571)272-4020. The examiner can normally be reached M-F 1000-1800 EST.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Bhavesh Mehta can be reached at 571-272-7453. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
JOSEPH J. YAMAMOTO
Examiner
Art Unit 2656
/BHAVESH M MEHTA/Supervisory Patent Examiner, Art Unit 2656
Prosecution Insights