DETAILED ACTION
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 .
This action is responsive to the amendments and remarks filed July 23, 2026, and the applicant-initiated interview of June 11, 2026. Claims 1-20 are pending, claims 1, 2, 11, and 17 are amended. Claims 3-10, 12-16, and 18-20 are not amended.
Response to Arguments
Applicant's arguments filed July 23, 2026 have been fully considered but they are not persuasive.
The applicant’s procedural arguments regarding the § 101 rejection have been considered. The Step 2A, Prong 2 analysis has been rewritten below to address the claim as a whole, including the additional elements and their interactions. The rejection is maintained on the rewritten analysis.
Applicant argues the claims provide a technical solution to a technical problem citing ¶¶3-4, 47-48, and Table 3 (¶¶212-214). The specification describes the problem with contrastive image-text scoring is that it does not capture word order and cannot model the co-dependency objects and attributes (¶¶174-175). The solution is generative prompting, in which the image-text model scores a prompt token by token, conditioned on the image and the preceding prompt tokens (¶¶ 186-187). The claims do not recite that solution. They recite processing image data and “prompts with a pre-trained image-text model to determine a probability score.” The specification states that contrastive scoring fails even for prompts that name both the object and attribute, such as “bell shaped sky” (¶ 175). Table 3 reports results for “the method which finetunes the generative prompts” (¶212), which is also not claimed.
Applicant cites Desjardins. In Desjardins, the claims recited a particular way of training a machine learning model, and specification tied that training to an improvement in how the model itself operates. Claims 1 and 11 recite no training, and claim 17 mentions captions only generically. The claims selects the inputs for two unmodified models and choose among their outputs.
Applicant’s argument that OVARNET fails to teach obtaining a plurality of prompt templates carrying object tokens and candidate attribute terms is not persuasive. OVARNET at p. 7 teaches, “using carefully designed prompt [9, 15], for example, ‘It is a photo of [category]’ and ‘The attribute of the object is [attribute]’ for the category and attribute words, indeed delivers improvements.” OVARNET at p. 9 further states, “’Attribute prompt’ means the prompt designed with formats similar to ‘A photo of something that is [attribute]’, while 'object-attribute prompt' denotes ‘A photo of [category] [attribute]’.” OVARNET therefore obtains multiple prompt templates, each carrying a category token and an attribute token.
Therefore, the argued limitations were written broad such that they read upon the cited references or are shown explicitly by the references. As a result, the claims stand as follows.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. The claim(s) recite(s) mental processes: determining candidate words that describe an object, filling those words into prompt templates, and selecting the attribute best supported by likelihood scores. This judicial exception is not integrated into a practical application because the additional elements are a generic computer system and storage media, data gathering, a language model, and a pre-trained image-text model recited only by their inputs and outputs, and in claim 17, generic caption training. They amount to mere instructions to apply the exception using generic computer and machine-learning components. The claims end at the determination and not use to control a machine or system. The claims terminate at a determination of an attribute value with not downstream machine control, transformation, or other technical application of the result.
Step 1 – Statutory Category.
All independent claims qualify. Claim 1 is a process, claim 11 is a machine, and claim 17 is an article of manufacture (CRM).
Step 2A, Prong 1 – Judicial Exception.
Claims 1, 1 and 17 recite mental processes that can practically be performed in the human mind or with pen and paper (MPEP 2106.04(a)(2)(III)): (a) determining “a plurality of candidate attributes, wherein the plurality of candidate attributes comprise attributes predicted to be candidate terms that describe attributes of the particular object” (given "cat," a person can list "calico," "fluffy" or "sleeping"; see (¶ 3); (b) “generating … a plurality of prompts based on the candidate attribute and the plurality 'of prompt plates" (filling templates such as "{att} {obj}" and "{obj} is {att}" (¶ 120)) and (c) "determining … a particular attribute ... based on the plurality of probability scores" (comparing scores and picking the best candidate). ‘Processing "with a language model" and "with a pre-trained image-text model" cannot practically be performed mentally and is evaluated below as additional elements.
Step 2A Prong 1: Yes, judicial exception recited.
Step 2A, Prong 2 – Practical Application.
The additional elements are: (1) the processors and computer-readable media; (2) obtaining image data, text data; an image and the prompt templates; (3) processing the text data with a language model; (4) processing the image data and prompts with a pre-trained image-text model to determine a probability score; (5) in claim 11, processing the image with the image-text model to generate text data; and (6) in claim 17, training the image-text model on image-caption pairs. Element (1) is generic computer components (MPEP 2106.05(f)). Element (2) is data gathering (MPEP 2106.05(g)). Elements (3)-(5) recite each model only by its input and output, not how the result is achieved , which is a mere instruction to apply the exception (MPEP 2106.05(f)(1); ¶ 144). Considered together, the language model supplies candidate terms, the templates format them, and the image-text model scores the text against the image. No step changes how either model is built, trained or executed, and the determined attribute is not used to control any machine or system.
Step 2A, Prong 2, claims do not integrate the exception into a practical application.
Step 2B - Inventive Concept (WURC analysis).
The "computing system comprising one or more processors" is a WURC (generic processor). The "language model" used to predict word sequences is a WURC per ¶¶[0039], [0144] which describes general NLP models. "Pre-trained image-text model" are WURC per spec at ¶[0173] that acknowledges CLIP, ALIGN as well-known large-scale image-text models. CoCa is cited as prior art foundational model. The "probability score" computation is a WURC mathematical operation. The combination of the sequential application of two known ML models (LM + image-text model) to attribute recognition is a routine combination in the field per admitted prior art at ¶¶[0172]-[0175].
Step 2B: No inventive concept. Claims 1, 11, 17 ineligible.
Dependent claim analysis.
Claims 2 and 15 are directed to prompt generation fed into image-text model. This is an additional data manipulation step, still abstract, no new technical improvement. Claims 3 and 5 are directed to a LM trained to predict word sequences. This describes a training methodology, WURC. Claim 16 and 20 recite an image encoder + unimodal + multi modal text decoders. This is the architectural recitation of Coca model per ¶¶[0130], [0195]; WURC per admitted prior art. Claims 7, 8, 9, 10, 13, and 14 recite attribute type specifications (color, texture, action,
specialization). These are field of use/data content limitations, and do not confer eligibility. Claim 17 (independent CRM) has same abstract idea as method/system.
No claim is currently eligible. The strongest path to eligibility would require amending to
recite specific architectural elements of the image-text model that achieve the improvement (e.g., the prefix LM cross-attention mechanism specifically tied to the attribute scoring) or adding downstream application of the determined attribute (e.g., to control display, annotation, or retrieval system operation).
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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1-3 and 5-10 is/are rejected under 35 U.S.C. 103 as being unpatentable over Chen et al., "OvarNet: Towards Open-Vocabulary Object Attribute Recognition," (hereinafter "OVARNET") in view of Chowdhery et al., "PaLM: Scaling Language Modeling with Pathways," (hereinafter "PALM").
Claim 1.
OVARNET and PALM disclose a computer-implemented method for attribute captioning, the method comprising:
obtaining, by a computing system comprising one or more processors, image data and text data, wherein the image data is descriptive of one or more objects, and wherein the text data is descriptive of a particular object associated with the image data (OVARNET: "given a training dataset … I … refers to an image, and yi = {bi, ci, ai} denotes its corresponding ground-truth annotations, with the coordinates for n object bounding boxes ... their corresponding semantic categories" (Sec. 3.1). This teaches obtaining image data depicting objects and text data (semantic category labels) descriptive of a particular object.);
obtaining, by the computing system, a plurality of prompt templates configured to include tokens associated with the particular object and candidate terms associated with one or more of the plurality of candidate attributes (OVARNETp. 9: “’Attribute prompt’ means the prompt designed with formats similar to ‘A photo of something that is [attribute]’, while 'object-attribute prompt' denotes ‘A photo of [category] [attribute]’.” OVARNET therefore obtains multiple prompt templates, each carrying a category token and an attribute token. OVARNET p. 7: “using carefully designed prompt [9, 15], for example, ‘It is a photo of [category]’ and ‘The attribute of the object is [attribute]’ for the category and attribute words, indeed delivers improvements.”);
for each of the plurality of candidate attributes:
generating, by the computing system, a plurality of prompts based on the candidate attribute and the plurality of prompt templates (Each is a fixed text frame with a slot for the object category and a slot for the attribute term. OVARNET p. 4: “we augment it with multiple learnable prompt vectors” (Sec. 3.2.2, Eq. 1).);
processing, by the computing system, the image data, and the plurality of prompts with a pre-trained image-text model to determine a probability score for the candidate attribute, wherein the probability score is descriptive of a likelihood the candidate attribute is associated with the image data (OVARNET: "Attribute prediction can be obtained by computing the similarity between visual region feature and attribute concept embedding as:
PNG
media_image1.png
30
361
media_image1.png
Greyscale
" (Sec. 3.2.2, Eq. 2). The sigmoid-normalized similarity score
s
^
ij is a probability score representing the likelihood that the ith image region contains the jth attribute. The pre-trained CLIP model serves as the pre-trained image-text model.); and
determining, by the computing system, a particular attribute of the plurality of candidate attributes is associated with the particular object depicted in the image data based on the plurality of probability scores (OVARNET: Attributes with confidence scores higher than 0. 7 are selected as positive (Sec. 3.2.3 Step-II). This teaches determining a particular attribute
based on the plurality of probability scores.).
OVARNET discloses all of the subject matter as described above except for specifically teaching “processing, by the computing system, the text data with a language model to determine a plurality of candidate attributes, wherein the plurality of candidate attributes comprise attributes predicted to be candidate terms that describe attributes of the particular object.” OVARNET uses a fixed attribute vocabulary of 620 attributes from the VAW dataset rather than generating candidates with a language model. However, PALM in the same field of endeavor teaches processing, by the computing system, the text data with a language model to determine a plurality of candidate attributes, wherein the plurality of candidate attributes comprise attributes predicted to be candidate terms that describe attributes of the particular object (PALM: "trained a 540 billion parameter, densely activated, autoregressive Transformer on 780 billion tokens of high-quality text" (Sec. 1 ). Given a text prefix such as an object name (e.g., "a cat that is"), PaLM predicts likely continuation tokens through autoregressive next token prediction, generating descriptive attribute terms ( e.g., "fluffy ," "calico," "small"). PaLM demonstrates "state-of-the-art few-shot learning results on hundreds of language understanding and generation benchmarks" (Abstract). This teaches using a language model to determine candidate attributes from text data.).
The proposed modification is additive, not substitutive. Specifically, PALM generates candidate attribute terms as raw text strings (e.g., "fluffy," "orange," "domestic"). These text strings are then tokenized and formatted using OVARNET's existing learnable prompt template
before being processed by CLIP's text encoder to produce attribute embeddings. OVARNET's CLIP-based similarity scoring (Eq. 2), BCE training loss (Eq. 3), and the 0.7 confidence threshold for attribute selection all remain unchanged. The only modification is the source of the attribute term strings fed into the tokenizer g(.), expanded from a static VAW dictionary lookup to PALM's autoregressive output. OVARNET's prompt vectors, text encoder, visual encoder, and similarity scoring operate identically regardless of how the input attribute terms were generated.
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to use PALM to generate candidate attribute terms for input to OVARNET's attribute scoring framework. OVARNET itself identifies the central problem as achieving "open-vocabulary" attribute recognition (Title) but is constrained by the fixed 620-attribute vocabulary of the VAW dataset (Sec. 4.1). A person of ordinary skill would have recognized that an autoregressive language model like PALM, which encodes statistical cooccurrence patterns between objects and their descriptive attributes across 780 billion tokens of natural language text, provides a direct solution to this vocabulary limitation. The combination applies a known technique (LM-based text generation) to a known problem (fixed-vocabulary constraint in attribute recognition) to yield predictable results (dynamically generated, object relevant candidate attributes that can be scored against image regions using OVARNET's existing similarity metric). KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007).
Claim 2.
OVARNET and PALM disclose the method of claim 1, further comprising: generating, by the computing system, the plurality of prompts with a plurality of different configurations based on the text data and the plurality of prompt templates, and the plurality of candidate attributes (OVARNET p. 9: ” Attribute prompt' means the prompt designed with formats similar to "A photo of something that is [attribute]", while 'object-attribute prompt' denotes "A photo of [category] [attribute]". For the 'combined prompt', the outputs of the 'attribute prompt' and the 'object-attribute prompt' are weighted average.”).
Claim 3.
OVARNET and PALM disclose the method of claim 1, wherein the language model was trained to predict word sequences (PALM: "we trained a 540-billion parameter, densely activated,
Transformer language model" (Abstract) "on 780 billion tokens of high-quality text" (Sec.
1). PaLM is an autoregressive language model explicitly trained to predict word sequences.
This teaches a language model trained to predict word sequences.).
Claim 5.
OVARNET and PALM disclose the method of claim 1, wherein the plurality of candidate attributes are determined based on learned word sequences, wherein the learned word sequences were learned by training the language model (PALM: PaLM's autoregressive training on 780 billion tokens learns statistical word sequence patterns. When given a text prefix describing an object, PaLM's predictions of likely attribute terms reflect these learned word
sequences. This teaches candidate attributes determined from learned word sequences
acquired during language model training.).
Claim 6.
OVARNET and PALM disclose the method of claim 1, wherein the pre-trained image-text model was trained on a training dataset comprising a plurality of training images and a plurality of training captions, wherein each of the plurality of training captions are descriptive of a respective caption for one or more of the plurality of training images (OVARNET: Sec. 3.2.3
Step-II: "we also consider using freely available image-caption datasets, e.g., Dimg-cap
{{I1, s1}, ... , {IN, sN}}, where the li, si refer to image and caption sentence respectively." OVARNET trains on CC-3M (3 million image-caption pairs) and COCO Cap (120k images with caption annotations). This teaches training the image-text model on training images with corresponding training captions.).
Claim 7.
OVARNET and PALM disclose the method of claim 1, wherein the particular attribute comprises a particular color (OVARNET Sec. 4.1: VAW includes color attributes. Sec. 3.2.3: "green" parsed as attribute from caption.).
Claim 8.
OVARNET and PALM disclose the method of claim 1, wherein the particular attribute comprises a particular texture for the particular object (OVARNET Sec. 4.1: VAW includes texture attributes. Sec. 3.2.3: "striped" parsed as attribute from "A striped zebra.").
Claim 9.
OVARNET and PALM disclose the method of claim 1, wherein the particular attribute comprises an action description for the particular object, wherein the action description is descriptive of an action being performed by the particular object in the image data (OVARNET Sec. 4.1: VAW has "620 attributes, for example, color, material, shape, size, texture, action." "Eating" parsed from "A striped zebra is eating green grass.").
Claim 10.
OVARNET and PALM disclose the method of claim 1, wherein the particular attribute comprises a specialization classification, wherein the specialization classification is descriptive of an object-specific adjective associated with the particular object ((OVARNET Sec. 3.2.2: ontology-aware prompts, e.g., "<I> clip-t(wet, state)" distinguishes object-specific attribute contexts.).
Claim 4, 11-14, and 16-20 are rejected under 35 U.S.C. 103 as being unpatentable over OVARNET in view of PALM, further in view of Yu et al. , "CoCa: Contrastive Captioners are Image-Text Foundation Models” (hereinafter "COCA").
Claim 4.
OVARNET and PALM disclose the method of claim 1, wherein the pre-trained image-text model was
OVARNET's CLIP is not trained for captioning, COCA teaches “trained to generate text captions for images, wherein the text captions are descriptive of features depicted in the image” (COCA Abstract: "captioning loss on the multimodal decoder outputs which predicts text tokens autoregressively.").
It would have been obvious to use COCA as the image-text model to improve the system by adding caption generation capability while retaining contrastive alignment for scoring, since COCA "subsum[es] model capabilities from contrastive approaches like CLIP" (Abstract). Adapting prompt vectors to COCA's encoder is a routine improvement as disclosed by Zhou’s "Learning to Prompt" and Yao’s "CPT" cited on the IDS.
Claim 11.
OVARNET and PALM disclose a computing system for language model conditioned image captioning, the system comprising: one or more processors; and one or more non-transitory computer-readable media that collectively store instructions that, when executed by the one or more processors, cause the computing system to perform operations (OVARNET: GPUbased
training/inference system), the operations comprising: obtaining an image, wherein the image is descriptive of one or more objects (OVARNET Sec. 3.1);
processing the text data with a language model to determine a plurality of candidate attributes, wherein the plurality of candidate attributes comprise attributes predicted to be candidate terms that describe attributes of the particular object (PALM: autoregressive prediction of attribute terms from generated caption text – see explanation and motivation to combine above in claim 1); for each of the plurality of candidate attributes: processing the image, text data, and candidate attribute with the pre-trained image-text model to determine a probability score for the candidate attribute, wherein the probability score is descriptive of a likelihood the candidate attribute is depicted in the image (OVARNET Eq. 2
PNG
media_image1.png
30
361
media_image1.png
Greyscale
); and determining a particular attribute of the plurality of candidate attributes is associated with the particular object depicted in the image based on the plurality of probability scores (OVARNET: confidence threshold selection).
OVARNET and PALM do not teach “processing the image with a pre-trained image-text model to generate text data, wherein the text data is descriptive of a particular object depicted in the image.” COCA teaches this via its multimodal decoder (COCA Abstract: "omits cross-attention in the first half of decoder layers to encode unimodal text representations, and cascades the remaining decoder layers which cross-attend to the image encoder for multimodal image-text representations.") OVARNET's training-time use of ground-truth labels (Sec. 3.2.3) does not preclude inference-time combination: its class-agnostic RPN (Sec. 3.2.1) generates proposals independently, and COCA's captions supply object text.
It would have been obvious to combine the three references to improve attribute recognition by enabling fully automatic operation without manual annotations. OVARNET's Step-II training already parses captions into categories and attributes (Sec. 3.2.3: "we use TextBlob to parse all captions into 'semantic category' , 'attribute' , and 'noun phrases"'). Extending this caption-to-attribute workflow from training to inference-using COCA's captions and PALM's predictions improves deployment efficiency by eliminating the annotation dependency.
Claim 12.
OVARNET, PALM, and COCA disclose the system of claim 11, wherein the plurality of candidate attributes comprise a plurality of terms determined to be associated with the particular object based on one or more learned sequences (PALM: autoregressive training learns object-attribute sequence patterns.).
Claim 13.
OVARNET and PALM disclose the system of claim 11, wherein the plurality of candidate attributes comprise a plurality of adjectives and a plurality of verbs (OVARNET Sec. 4.1: VAW's 620 attributes include adjective-type (e.g., "red," "large") and verb/action-type ( e.g., "eating," "standing").).
Claim 14.
OVARNET and PALM disclose the system of claim 11, wherein the plurality of candidate attributes comprise one or more color attributes and one or more texture attributes (OVARNET
Sec. 4.1: VAW includes "color ... texture." Fig. 2: "white, color" as attribute category; "striped" parsed from captions.).
Claim 16.
OVARNET and PALM disclose the system of claim 11, wherein the operations further comprise: before obtaining the image: obtaining a training dataset, wherein the training dataset comprises a plurality of training examples, wherein each training example comprises an image example and a respective caption example, wherein the respective caption example is descriptive of a caption for the image example; and training an image-text model based on the training dataset to generate captions for input images (OVARNET Sec. 3.2.3 Step-II: trains on CC-3M and COCO-Cap. COCA Abstract: pretrained "on both web-scale alt-text data and annotated images.").
Claim 17.
OVARNET, PALM, and COCA disclose one or more non-transitory computer-readable media that collectively store instructions that, when executed by one or more computing devices, cause the one or more computing devices to perform operations, the operations comprising: obtaining a training dataset, wherein the training dataset comprises a plurality of training examples, wherein each training example comprises an image example and a respective caption example, wherein the respective caption example is descriptive of a caption for the image example (OVARNET Step-II: CC-3M, COCO-Cap); training an image-text model based on the training dataset to generate captions for input images (COCA Abstract: "captioning loss on the multimodal decoder outputs which predicts text tokens autoregressively."); obtaining image data and text data, wherein the image data is descriptive of one or more objects, and wherein the text data is descriptive of a particular object associated with the image data (OVARNET Sec. 3.1); processing the text data with a language model to determine a plurality of candidate attributes, wherein the plurality of candidate attributes comprise attributes predicted to be candidate terms that describe attributes of the particular object (PALM: autoregressive prediction of candidate terms); for each of the plurality of candidate attributes: processing the image data, text data, and candidate attribute with the image-text model to determine a probability score for the candidate attribute, wherein the probability score is descriptive of a likelihood the candidate attribute is associated with the image data (OVARNET Eq. 2); and determining a particular attribute of the plurality of candidate attributes is associated with the particular object depicted in the image data based on the plurality of probability scores (OVARNET: threshold selection). The motivations to combine above apply to this claim, mutatis mutandis.
Claim 18.
OVARNET, PALM, and COCA disclose the one or more non-transitory computer-readable media of claim 17, wherein the text data is descriptive of the particular object and a particular adjective for the particular object (OVARNET Sec. 3.2.3: noun phrases "striped zebra," "yellow frisbee.").
Claim 19.
OVARNET, PALM, and COCA disclose the one or more non-transitory computer-readable media of claim 18, wherein the plurality of candidate attributes are determined based on a text string comprising the particular object and the particular adjective (PALM: given "striped zebra," predicts additional attributes like "African," "wild," "grazing.").
Claim 20.
OVARNET, PALM, and COCA disclose the one or more non-transitory computer-readable media of claim 17, wherein the image-text model comprises: one or more image encoders; one or more unimodal text decoders; and one or more multimodal text decoders (COCA Abstract: "omits cross-attention in the first half of decoder layers to encode unimodal text representations, and cascades the remaining decoder layers which cross-attend to the image encoder for multimodal image-text representations." This is exactly: (1) image encoder, (2) unimodal text decoder, (3) multimodal text decoder.) It would have been obvious to select COCA' s architecture to improve both scoring accuracy and captioning quality in a single model, since it is the known embodiment of this three-component design optimized for both contrastive alignment and generative captioning, per KSR (finite identified predictable solutions).
Allowable Subject Matter
Claim 15 is 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, and overcoming the §101 rejection.
Claim 15 requires iteratively feeding a determined attribute back into the LM to generate additional candidate attributes, then re-scoring those additional candidates with the image-text model. This autoregressive attribute refinement chain is not taught or suggested by OVARNET, PALM, COCA, or any other reference of record.
Conclusion
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Ross Varndell whose telephone number is (571)270-1922. The examiner can normally be reached M-F, 9-5 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, O’Neal Mistry can be reached at (313)446-4912. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see https://ppair-my.uspto.gov/pair/PrivatePair. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/Ross Varndell/Primary Examiner, Art Unit 2674
Prosecution Insights