DETAILED ACTION
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 08/13/2026 has been entered.
The application contains claims 1-4, 6-12, 14-20, 22-25, all examined and rejected.
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 .
Response to Amendment
It is acknowledged that claims 1, 4, 7, 8, 9, 12, 15, 16, 1, 20, 23 and 24 were amended. Claims 5, 13 and 21 were canceled.
Response to Arguments
Applicant’s arguments with respect to the amended claim(s) 1, 9 and 17 have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
Double Patenting
The nonstatutory double patenting rejection is based on a judicially created doctrine grounded in public policy (a policy reflected in the statute) so as to prevent the unjustified or improper timewise extension of the “right to exclude” granted by a patent and to prevent possible harassment by multiple assignees. A nonstatutory double patenting rejection is appropriate where the conflicting claims are not identical, but at least one examined application claim is not patentably distinct from the reference claim(s) because the examined application claim is either anticipated by, or would have been obvious over, the reference claim(s). See, e.g., In re Berg, 140 F.3d 1428, 46 USPQ2d 1226 (Fed. Cir. 1998); In re Goodman, 11 F.3d 1046, 29 USPQ2d 2010 (Fed. Cir. 1993); In re Longi, 759 F.2d 887, 225 USPQ 645 (Fed. Cir. 1985); In re Van Ornum, 686 F.2d 937, 214 USPQ 761 (CCPA 1982); In re Vogel, 422 F.2d 438, 164 USPQ 619 (CCPA 1970); In re Thorington, 418 F.2d 528, 163 USPQ 644 (CCPA 1969).
A timely filed terminal disclaimer in compliance with 37 CFR 1.321(c) or 1.321(d) may be used to overcome an actual or provisional rejection based on nonstatutory double patenting provided the reference application or patent either is shown to be commonly owned with the examined application, or claims an invention made as a result of activities undertaken within the scope of a joint research agreement. See MPEP § 717.02 for applications subject to examination under the first inventor to file provisions of the AIA as explained in MPEP § 2159. See MPEP § 2146 et seq. for applications not subject to examination under the first inventor to file provisions of the AIA . A terminal disclaimer must be signed in compliance with 37 CFR 1.321(b).
The filing of a terminal disclaimer by itself is not a complete reply to a nonstatutory double patenting (NSDP) rejection. A complete reply requires that the terminal disclaimer be accompanied by a reply requesting reconsideration of the prior Office action. Even where the NSDP rejection is provisional the reply must be complete. See MPEP § 804, subsection I.B.1. For a reply to a non-final Office action, see 37 CFR 1.111(a). For a reply to final Office action, see 37 CFR 1.113(c). A request for reconsideration while not provided for in 37 CFR 1.113(c) may be filed after final for consideration. See MPEP §§ 706.07(e) and 714.13.
The USPTO Internet website contains terminal disclaimer forms which may be used. Please visit www.uspto.gov/patent/patents-forms. The actual filing date of the application in which the form is filed determines what form (e.g., PTO/SB/25, PTO/SB/26, PTO/AIA /25, or PTO/AIA /26) should be used. A web-based eTerminal Disclaimer may be filled out completely online using web-screens. An eTerminal Disclaimer that meets all requirements is auto-processed and approved immediately upon submission. For more information about eTerminal Disclaimers, refer to www.uspto.gov/patents/apply/applying-online/eterminal-disclaimer.
Claims 1-4, 6-12, 14-20, 22-25 rejected on the ground of nonstatutory double patenting as being unpatentable over claims 1, 4, 5 and 6 of U.S. Patent No. 12737425. Although the claims at issue are not identical, they are not patentably distinct from each other because the limitation from 12737425 fully anticipate the claims from the instant Application see mapping below.
Instant Application 19/038072
U.S. Patent No. 12737425
Claims 1, 9 and 17 recite:
… generating a query vector based at least in part on a query received from a client computing device;
generating a user account vector different from the query vector, wherein the user account vector is generated based at least in part on a user account associated with the client computing device;
generating a combined vector by concatenating the query vector and the user account vector;
generating an output vector based at least in part on the combined vector, wherein the output vector is generated by processing the combined vector using a transformer-based large language model (LLM);
obtaining, based at least in part on the query, a plurality of item vectors, wherein each item vector of the plurality of item vectors corresponds to a respective item;
comparing the output vector to the plurality of item vectors to generate respective similarity scores;
ordering the plurality of item vectors in accordance with the similarity scores to establish an ordered plurality of item vectors;
and causing the client computing device to display, in accordance with the ordered plurality of item vectors, respective affordances for a plurality of items that correspond to the ordered plurality of item vectors.
Claim 1 recite: receiving a query, wherein the query is associated with a user account, and the user account is associated with a user account vector;
generating a query vector based at least in part on the query;
(from above and claim 4): a user account vector… the user account vector is generated based at least in part on… the user account
(from claim 5): concatenating the query vector to the user account vector
generating an output vector based at least in part on the query vector and the user account vector; (from claim 6): output vector is generated based at least in part on the query vector and the user account vector using a transformer-based large language model (LLM)
obtaining, based at least in part on the query, a plurality of digital asset vectors, wherein each digital asset vector of the plurality of digital asset vectors corresponds to a respective digital asset;
comparing the output vector to the plurality of digital asset vectors to generate respective similarity scores for the plurality of digital asset vectors;
filtering the plurality of digital asset vectors in accordance with the similarity scores to establish a filtered plurality of digital asset vectors;
and displaying, in accordance with the filtered plurality of digital asset vectors, respective affordances for the respective digital assets that correspond to the filtered plurality of digital asset vectors.
The corresponding dependent claims are rejected for the sabe reasons above as they depend on rejected independent claims 1, 9 and 17.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
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.
Claim(s) 1, 2, 3, 4, 6, 9, 10, 11, 12, 14, 17, 18, 19, 20, 22 and 25 is/are rejected under 35 U.S.C. 103 as being unpatentable over Pastor et al. (US 20190294690 A1) in view of Doddapaneni et al. "User embedding model for personalized language prompting." 10 Jan 2024), Heitz et al. (US 20120290621 A1) and Arora et al. (US 20210241343 A1).
Regarding claim 1, Pastor discloses:
a method for performing searches, the method comprising, by a server computing device: generating a query vector based at least in part on a query received from a client computing device, generating a user account vector different from the query vector, wherein the user account vector is generated based at least in part on a user account associated with the client computing device; generating a combined vector by concatenating the query vector and the user account vector; generating an output vector based at least in part on the combined vector, wherein the output vector is generated by processing the combined vector using a transformer-based large language model (LLM), at least by (paragraph [0150] where the taste vector associated with the user identifier is the recited user account is associated with a user account vector, paragraph [0153] further describes receiving “suggestion characteristic data 232 of the playlist character suggestions 230 that have been selected by the user” and paragraph [0154] describes representing the suggestion characteristic data as the seed vector, as shown above, the seed vector represents user selected playlist that the user is requested suggestions for (query) and also the user’s taste vector associated with the user account, so at very least, the seed vector represents the output vector where the seed vector represents the combination of user’s request and user account preferences, which is used to identify and recommend media content items based on the similar vectors. However, Pastor does not: (a) generating a query vector… (b) generating a user account vector different from the query vector… (c) generating a combined vector by concatenating the query vector and the user account vector; (d) and the seed vector is not generating an output vector based at least in part on the combined vector …generated by processing the combined vector using a transformer-based large language model (LLM);
obtaining, based at least in part on the query, a plurality of item vectors, wherein each item vector of the plurality of item vectors corresponds to a respective item, at least by (paragraph [0156] “vectors for other media content items… corresponding vectors include vectors that represent different types of data elements”)
comparing the output vector to the plurality of item vectors to generate respective similarity scores, at least by (paragraph [0156-0158] “compare the seed vector with other corresponding vectors… determines vectors that are similar to the seed vector… similarity or comparison measurement can be used to compare two vectors”; where the similarity or comparison measurement is the generated respective similarity scores)
at least by (paragraph [0159] “determines media content items based on the similar vectors”
and causing the client computing device to display, in accordance with the at least by (paragraph [0159-0160] “Once one or more similar vectors are identified at the operation 438, at least some of the media content items associated with the similar vectors can be identified and determined as recommended media content items for the current playlist…update the current playlist with at least one of the recommended media content items.”)
Pastor also fails to specifically describe: (e) ordering the plurality of item vectors, and displaying ordered items corresponding to the ordering of the item vectors
However, Doddapaneni teaches the above limitation(s):
(a) at least by (Sec. 2, “given a task instruction X, the LM embeds these tokens to a matrix Xe ∈ Rn× e” therefore the query/input is converted to an embedding representation (e.g. query vector)
(b) at least by (Sec. 2, “textual user history H = {hi}p i=1 is converted into embeddings U = {ui}p i=1 with SentenceT5 (Ni et al., 2022). Each history item ui is a composite of three distinct embeddings: (i) title & genre, (ii) rating, and (iii) description. The collective history of ‘p’ items is expressed as U ∈ Rp×3s, where s corresponds to the embedding dimension of SentenceT5. These embeddings undergo processing within a transformer network (UEM). To ensure dimension alignment with e, a linear projection layer is intro duced atop the transformer, mapping the dimension 3s to e, thereby yielding PrUEM(U) ∈ Rp× e” which describes user-history embedding U are processed by transformer UEM yielding PrUEM(U) (e.g. user history vector different from the query vector) (not user account vector)
(c) at least by (Sec. 2 “user and task prompts are concatenated with the input embedding, resulting in a unified embedding matrix, represented as [Pe; PrUEM(U);Xe] ∈ R(k+p+n)× e.”
(d) at least by (Sec. 2 futher describing after concatenation, “This composite embedding flows through the LM”, to produce Y)
And, Heitz teaches the above limitation (e) at least by (paragraph [0062-0063] “Using the resulting vectors (embeddings) in the embedding space, for a given input, it is possible to rank possible outputs of interest such that the highest-ranked outputs are the best semantic match for the given input. For example, the following ranking functions, f.sub.i(x), identify, for a given seed track, a number of songs that are similar to the seed track, and a number of artists who are similar to the artist who performed the seed track… ranked list of desired results (e.g., a ranked list of audio tracks that have auditory features that are similar to an auditory feature of a seed track”)
Furthermore, Arora teaches the modification from user history vectors to user account vectors at least by paragraph [0105] further describes generated “user embedding h.sub.u” where paragraph [0107-0110] further describes deep learning model generates “user embeddings 725” where the inference engine obtains the user embedding using a user identifier (user account) therefore the user embedding is the user account embedding
Therefore, before the effective filing date of the invention it would have been obvious to one of ordinary skill in the art to modify the seed vector of Pastor with Doddapaneni and Arora to provide the ability to generate the seed vector by using an LLM to process the concatenated query and user history/account embeddings to “handle significantly larger histories compared to traditional text-based approaches, resulting in improved predictive performance” (Doddapaneni, Sec. 5) and to “provide[ing] personalized recommendations through large-scale deep-embedding architecture” (Arora, para. 0029). It would have been obvious to one of ordinary skill in the art to combine Pastor with Heitz to provide a ranked list of audio tracks based on similarity calculations provided by Heitz to improve the user’s ability to access to more relevant content by “improv[ing] the sequencing of the playlist”, (Heitz, 0046).
As per claim 2, claim 1 is incorporated and Pastor further describes:
wherein the query comprises text content, image content, audio content, video content, or some combination thereof, at least by (paragraph [0090-0091, 0105] describes text content as “user can describe the playlist with a playlist title”, “words or phrases”; paragraph [0106] describes audio content as “media content attribute description 240 can include item description information and acoustic attribute information… identifies one or more media content item” therefore describes some combination of text and audio content, that are included into the suggestion characteristic data that forms the seed vector).
As per claim 3, claim 1 is incorporated and Pastor further describes:
wherein the query vector is generated based at least in part on the query at least by (paragraph [0154] describes representing the suggestion characteristic data as the seed vector (e.g. query vector)
But Pastor and Heitz fails to specifically recite using a transformer-based large language model (LLM) to generate the vector
However, Doddapaneni teaches the above limitation(s) at least by (Sec. 2, “given a task instruction X, the LM embeds these tokens to a matrix Xe ∈ Rn× e” therefore the query/input is converted to an embedding representation (e.g. query vector)
Therefore, before the effective filing date of the invention it would have been obvious to one of ordinary skill in the art to modify the seed vector of Pastor with Doddapaneni to provide the ability to generate the seed vector by using an LLM to process the concatenated query and user history/account embeddings to “handle significantly larger histories compared to traditional text-based approaches, resulting in improved predictive performance” (Doddapaneni, Sec. 5).
As per claim 4, claim 1 is incorporated and Pastor further describes:
wherein the user account vector is generated based at least in part on: a first set item vectors that correspond to items marked as favorites under the user account, at least by (paragraph [0078] which describes description vectors that represent metadata representing “lists of favorite songs”)
a second set of item vectors that correspond to Decisions included in a library of items associated with the user account, at least by (paragraph [0150-0151] describes vectors that represent user preferences and tastes)
a third set of item vectors that correspond to items accessed by the user account within a first threshold period of time, paragraph [0150] describes a taste vector associated with user identifier that the represents the user’s “long term” tasted which describes a first threshold period of time.
and a fourth set of query vectors that correspond to queries provided in association with the user account within a second threshold period of time, at least by (paragraph [0150] “the taste vector can define the user's real-time listening taste.” where real-time describes a second threshold period of time)
As per claim(s) 5, 13 and 21, canceled.
As per claim 6, claim 1 is incorporated and Pastor further describes:
wherein, and the transformer-based LLM implements a set of fully connected layers and a set of input normalization layers,at least by (paragraph [0154] the seed vector is the generated output vector based on the combined vector)
But Pastor and Heitz fails to specifically recite
However, Doddapaneni teaches the above limitation(s) at least by (Fig. 1 which describes the multi-head attention (e.g. fully connected layers) with Add & Norm (e.g. a set of input normalization layers, Sec. 3 further describes “3 transformer layers with 12 attention heads, 768d embeddings, and 2048d MLP layers”)
Therefore, before the effective filing date of the invention it would have been obvious to one of ordinary skill in the art to modify the seed vector of Pastor with Doddapaneni to provide the ability to generate the seed vector by using an LLM to process the concatenated query and user history/account embeddings to “handle significantly larger histories compared to traditional text-based approaches, resulting in improved predictive performance” (Doddapaneni, Sec. 5).
As per claim 25, claim 1 is incorporated and Pastor further describes:
wherein the output vector includes novel generative content as a result of processing the combined vector using the transformer-based LLM, at least by (paragraph [0154] describes representing the suggestion characteristic data as the seed vector, as the out put vector) but fails to describe the output vector being from processing the combined vector using the transformer-based LLM,
However, Doddapaneni describes the above limitation at least by (Sec. 2 futher describing after concatenation, “This composite embedding flows through the LM”, to produce Y)
Therefore, before the effective filing date of the invention it would have been obvious to one of ordinary skill in the art to modify the seed vector of Pastor with Doddapaneni to provide the ability to generate the seed vector by using an LLM to process the concatenated query and user history/account embeddings to “handle significantly larger histories compared to traditional text-based approaches, resulting in improved predictive performance” (Doddapaneni, Sec. 5).
Claim(s) 9, 10, 11, 12 and 14 recite equivalent claim limitations as claim(s) 1, 2, 3, 4 and 6 above, except that they set forth the claimed invention as a non-transitory computer readable storage medium; Claim(s) 17, 18, 19, 20 and 22 recite equivalent claim limitations as claim(s) 1, 2, 3 and 4 and 6 above, except that they set forth the claimed invention as a server computing device; as such they are rejected for the same reasons as applied hereinabove.
Claim(s) 7-8, 15-16, 23-24 is/are rejected under 35 U.S.C. 103 as being unpatentable over Pastor, Doddapaneni, Heitz and Arora further in view of Glesinger et al. (US 20240104305 A1).
As per claim 7, claim 1 is incorporated and Pastor further describes:
wherein a given item vector of the plurality of item vectors is generated by: obtaining, at least by (paragraph [0077] “Acoustic metadata may also include spectral information such as melody, pitch, harmony, timbre, chroma, loudness, vocalness, or other possible features. Acoustic metadata may take the form of one or more vectors”)
obtaining, at least by (paragraph [0078] “Cultural metadata may be derived from expert opinion such as music reviews or classification of music into genres… Cultural metadata may take the form of one or more vectors”
and generating the item vector based at least in part on combining the first and second item vectors, at least by (paragraph [0080 and 0156] describes the vector(s) representing Acoustic metadata and Cultural metadata representing a particular content item, and in identifying a similarity to a seed vector, the vectors for each media content item are compared, as such vectors for the media content item, is the item vector describes that combines the items acoustic metadata vector and cultural metadata vector)
But Pastor and Heitz fails to specifically recite
However, Glesinger teaches the above limitations at least by (paragraph [0168] “semantic chains are converted to numeric-based representations by a vector embedding process such as a process that applies trained neural networks (e.g., Large Language Models or LLMs)”; and paragraph [0189] “application of statistical-based methods such as neural networks, including long short-term memory (LSTM) deep learning neural networks and/or associated variations of LSTM such as Gated Recurrent Units (GRUs), or transformer-based models”)
Therefore, before the effective filing date of the invention it would have been obvious to one of ordinary skill in the art to combine the system of Pastor and Heitz with Glesinger to “search for subsets of the content that are most relevant to the one or more semantic chains” (e.g. combined vectors) through “application of statistical-based methods such as neural networks, including long short-term memory (LSTM) deep learning neural networks and/or associated variations of LSTM such as Gated Recurrent Units (GRUs), or transformer-based models”, (Glesinger, para. 0189); “improving the system's interpretations of content,” (Glesinger, para. 0199); and accuracy, “enabling increasingly complex and subtle inferences by the system,” (Glesinger, para. 0192).
As per claim 8, claim 7 is incorporated and Pastor further describes:
wherein, when the corresponding respective item comprises an audio file:the metadata comprises the following song properties: album, artist, title, track number, genre, year, duration, bitrate, sample rate, channels, composer, comment, copyright, encoder, language, publisher, original artist, album artist, disc number, lyrics, mood, tempo, key, ISRC (International Standard Recording Code),recording date, release date, label, BPM (beats per minute), performer, conductor, compilation, part of a set, podcast, podcast URL, podcast ID, podcast feed, episode number, episode ID, episode URL, cover art, custom tags, or some combination thereof, at least by (paragraph [0079] “ Explicit metadata may include album and song titles, artist and composer names, other credits, album cover art, publisher name and product number, and other information”
and the data content comprises the following song characteristics: melody, harmony, rhythm, tempo, meter, lyrics, chorus, verse, bridge, dynamics, instrumentation, arrangement, key, harmonic progression, timbre, form, texture, style, emotion, production, hook, groove, transition, or some combination thereof, at least by (paragraph [0077] “Acoustic metadata may include temporal information such as tempo, rhythm, beats, downbeats, tatums, patterns, sections, or other structures. Acoustic metadata may also include spectral information such as melody, pitch, harmony, timbre, chroma, loudness, vocalness, or other possible features”)
Claim(s) 15-16, recite equivalent claim limitations as claim(s) 6-8, above, except that they set forth the claimed invention as a non-transitory computer readable storage medium; Claim(s) 23-24, recite equivalent claim limitations as claim(s) 6-8, above, except that they set forth the claimed invention as a server computing device; as such they are rejected for the same reasons as applied hereinabove.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Baltescu et al. (US 20230252550 A1): paragraph [0031] “recommendation service 260-N may be configured to receive a text query (e.g., a learned representation that is representative of the text query, such as a text embedding, etc.), an image query (e.g., a learned representation that is representative of the content item with which the user interacted, such as an image embedding, etc.), and the like, and determine, identify, and/or retrieve recommended products based on the query.” Which describes text query -> text embedding (e.g. query embedding) and image query (content item with which the user interacted) -> image embedding (e.g. user interacted content/account) embedding. And further paragraph [0037] “ a linear transformation may be applied to each of the set of image embeddings and the set of text embeddings. The transformed embeddings may then be concatenated with a global token and passed through a transformer encoder.” Which describes claimed generating a combined vector by concatenating the query vector and the user account vector; generating an output vector based at least in part on the combined vector …generated by processing the combined vector using a transformer-based large language model (LLM);
Huang et al. (US 20230315781 A1): paragraph [0044] “The embedding generator 323 may then perform an element-wise product between the query item embedding 314 and the user embedding 322 to form the personalized item embedding 320.” Which describes claimed a combined vector by concatenating the query vector and the user account vector; generating an output vector based at least in part on the combined vector.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to DENNIS TRUONG whose telephone number is (571)270-3157. The examiner can normally be reached Monday - Friday 8:30 am - 5:30 pm PT.
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, Amy Ng can be reached at (571) 270-1698. 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.
/DENNIS TRUONG/Primary Examiner, Art Unit 2164 09/28/2026
