[00:00:00] Speaker 05:
Our next argument is 25-1306 and 25-1470, Spiro v. Mercedes-Benz. Mr. Angeleri. May I please the Court? My name is Frank Angeleri for Spiro and these Mercedes-Benz appeals.

[00:00:20] Speaker 01:
So these appeals are the results of 12 IPRs against the three patents at issue. And in the Mercedes case, we believe the board erred in three ways.

[00:00:31] Speaker 03:
Before you do that, can you just tell me whether or not you agree with the housekeeping take that I was able to hear? in the prior appeal in terms of the impact between the Mercedes-Benz IPRs and the Volkswagen ones?

[00:00:45] Speaker 01:
I think the short answer to your question, Your Honor, is yes. So the Mercedes are broader than the VW. There are differences, as Mr. Lurie pointed out, but with respect to the question, I think after reflecting and conferring, we agree.

[00:00:57] Speaker 03:
So if we were to affirm with respect to the final written decisions in the Mercedes-Benz appeals, it would moot the Volkswagen appeal?

[00:01:04] Speaker 01:
I believe that's correct, Your Honor.

[00:01:05] Speaker 03:
Thank you.

[00:01:06] Speaker 01:
Okay. So with respect to the errors, we believe the board basically, there was a gap between the claims and Alden. Alden's their primary reference. And they closed that gap by expanding the claims, expanding Alden, and expanding the petitioner's arguments.

[00:01:25] Speaker 01:
With respect to the claims, I think the clearest example is the classifying limitation.

[00:01:32] Speaker 01:
which is effectively under the board's construction eliminated from the claim.

[00:01:39] Speaker 05:
How so? I mean, classifying the object, the board said it meant arranging or grouping the object according to some system or principle. Kind of sounds plain meaning-ish to me. So how is that inconsistent with the claim?

[00:01:53] Speaker 01:
So there's two points on that, Your Honor. The language that you just read is the Board's sort of refashioned version of the ordinary meaning dictionary definition that we put into evidence.

[00:02:03] Speaker 01:
The ordinary meaning definition, which the Board started with, the dictionary that the Board started with, uses the word classes. So it says arranging or grouping classes. in classes according to some system or principle. So they omitted the word classes, which in our view is very important because in order to classify from a plain meaning perspective, you need multiple classes. Otherwise, there's nothing to classify. Everything goes in the same.

[00:02:29] Speaker 03:
What is your best intrinsic evidence for it that the board incorrectly construed classifying?

[00:02:34] Speaker 01:
I'm sorry.

[00:02:35] Speaker 03:
What is your best intrinsic evidence support that the board incorrectly construed classifying?

[00:02:41] Speaker 01:
Absolutely. It's the claim. So it would be limitation five, Your Honor.

[00:02:46] Speaker 01:
The claim is, do you have claim 48 in front of you? It's in the blue brief, page 36, if you need to find it.

[00:02:53] Speaker 03:
Okay.

[00:02:57] Speaker 03:
And... So what you've identified as limitation five on page 36 of the blue brief?

[00:03:03] Speaker 01:
Yes. Do you see that? I do. Yeah. So that breakdown, by the way, is the breakdown that they use in their petition, which is Appendix Page 5229.

[00:03:11] Speaker 01:
You don't need to look at that.

[00:03:13] Speaker 03:
I'm just saying that's... And just for record purposes, right now I'm looking at what I'm going to say is Appeal 1306 is the lead. Is that correct?

[00:03:18] Speaker 01:
That's correct, Your Honor.

[00:03:21] Speaker 05:
Okay, so now link it up for me. Why does element five somehow at odds with the board's determination of what the word classifying means?

[00:03:31] Speaker 01:
Will do.

[00:03:33] Speaker 01:
Limitation one talks about an object indicated by data from one or more sensors. So you have a sensed object.

[00:03:40] Speaker 01:
Limitation five says that the first light is determined based on, based at least in part on classifying the object. If you have only one class, every time you sense it, you will light it in a certain way. You are no longer, the first light, if you have only one class, the first light is not determined as the claim requires.

[00:04:03] Speaker 01:
based on classifying the object. Rather, it's determined based on sensing the object, which means that limitation is essentially read out of the claim.

[00:04:12] Speaker 01:
The plain meaning of classifying, per the dictionary definition the board decided, has classes. This claim limitation, in order to be operative, requires classes. And so for that reason, Your Honor, we believe that the board's construction effectively read out not only the class is part of the ordinary meaning, but essentially the classifying limitation, limitation five, out of the claim. Do you have any other questions on classifying?

[00:04:46] Speaker 05:
Well, I mean, some of the embodiments seem to, as you're suggesting, distinguish objects by type, but others don't. So that's kind of where I'm wondering whether or not this patent really does require the identification of multiple classes that these different objects would be put into. Yeah.

[00:05:10] Speaker 01:
I don't know if the patent requires it. It's not a position that the patent requires it, Your Honor. We're not saying that the specification demands it. We're saying that this claim requires it. And so the patent has examples perhaps that involve classifying within classes and perhaps other examples that don't. But this claim would focus on the examples where you are, you know, classifying an object within multiple classes, within one of multiple classes.

[00:05:35] Speaker 05:
So the specification for us is consistent with our interpretation in the sense that... Why don't you show me in the specification itself, because I think that different embodiments use the term identifying and different embodiments use the term classifying. And so why don't you show me in the specification itself the one or more of the embodiments that you think are using the term classifying in a way that makes it clear that it requires multiple... distinct classes to choose from which to put the object in?

[00:06:09] Speaker 01:
I don't know that it uses the term classifying. The abstract talks about optimal lighting for task object. That's Appendix 49.

[00:06:19] Speaker 05:
Is there a column and line number you can put in?

[00:06:23] Speaker 01:
That's the abstract. So it's Appendix 49.

[00:06:30] Speaker 01:
They talk about Maximizing aiming and illumination level, that's Appendix 99 at Column 50, lines 50 to 58. And then the Board's decision on the identifying... Column 50?

[00:06:46] Speaker 05:
I'm sorry, Column 50, what line number?

[00:06:48] Speaker 01:
50 to 58. That's optimizing aiming and illumination level.

[00:06:53] Speaker 05:
I guess I'm just kind of lost as to how this helps inform...

[00:07:00] Speaker 05:
your argument that the Board got the definition of classifying wrong?

[00:07:06] Speaker 01:
Well, it informs it only in the sense that the specification, I think actually the specification uses identifying, which the Board itself cites in discussing identifying, for many of these situations where they are in fact deciding whether a sensed object fits in, you know, one of multiple different categories.

[00:07:26] Speaker 01:
Having said that, I think that the board, rather the specifications use of identifying to... Well, can I just get clear? Sure. Does the specification use the word classified? It uses it a few times, but not a ton. So we're not really relying on it.

[00:07:41] Speaker 05:
Show me where.

[00:07:41] Speaker 01:
Sure.

[00:07:56] Speaker 01:
It talks about classification on column 45, line 2. It talks about to classify or locate the contents.

[00:08:11] Speaker 01:
Again, it uses identifying more for these sort of separations.

[00:08:15] Speaker 02:
And are you contending that there is some of the use of the term classifying in the spec informs us? That your definition rather, your construction rather than the boards is correct? Am I contending that the use of classifying the spec? Yeah.

[00:08:28] Speaker 01:
I mean, are you relying on that language in the spec? No, I'm not relying on the language classifying the spec. I'm relying on the fact that the specification teaches the concept of sensing an object and then deciding does that object fit in this class or this class, and then I treat it differently.

[00:08:48] Speaker 03:
It seems like, Counsel, you're just relying on what you said was the fifth limitation of the claim itself, nothing specifically elsewhere in the specification and a dictionary definition to support up your proposed construction.

[00:09:04] Speaker 01:
Is that right? I think that's fair that we rely on limitation five and the argument I just made about how it's redundant if you just...

[00:09:15] Speaker 01:
The argument is that limitation five, you are no longer determining light based on a classification. You're determining light based on sensing, which takes that out. That's the limitation. The specification examples that we put in our brief that deal with the idea of selective illumination, which are simply consistent, but we're not saying it limits. And then the ordinary meaning, which, of course... As I mentioned, the district court cited our dictionary, but then altered it. Excuse me, the board, not district court, cited our dictionary. But those are the three areas of information we're relying on, Your Honor.

[00:09:46] Speaker 01:
Yes.

[00:09:47] Speaker 05:
Do you want to touch on some of the other issues?

[00:09:49] Speaker 01:
Sure.

[00:09:52] Speaker 01:
The Alden issue, I think, is briefed in the sense that we just don't think the board, we think the board erred when it said that Alden teaches LEDs that are off unless turned on, but that's heavily briefed. I would just come over. I think there are three areas where the board expanded petitioners' petition, and that's the argument we made on Claim 9, Claim 21, and Claim 26, Claim 26 of the 503 patent.

[00:10:24] Speaker 01:
But Claim 9 in particular, for example, you have a situation where the claim requires This is, it's on, the language of the claim is reproduced, I think, compliments to Ben's bread brief at 58. It has probably the best language for the claim.

[00:10:45] Speaker 01:
But if you distill it down, the language that, or the requirements that we are focusing on, claim nine requires control to diminish glare based on a camera and an on-camera.

[00:11:05] Speaker 01:
And the petitioners went through three different theories. Well, they went through two, and then the board came up, in our view, with a third to try to find a theory where Alden controls to diminish glare based on a non-camera sensor.

[00:11:21] Speaker 01:
And none of the theories teach that.

[00:11:25] Speaker 01:
The board's theory that it ultimately adopted is different from those that were raised by the petitioner.

[00:11:33] Speaker 01:
We know that because they cited our surreply, not the petitioner or the reply.

[00:11:39] Speaker 01:
We know that because the theory is different. The first two theories that Mercedes raised, the other vehicle is already in the scene before the turn, and they're talking about how they're going to manage that. In the third theory, the vehicle arrives into the scene after the turn. And I say that for the proposition that the board, in our view, did overreach on Claim 9 by articulating a theory that wasn't in the petition and the reply.

[00:12:09] Speaker 01:
And I'll leave the rest to the briefs unless the court has questions.

[00:12:13] Speaker 05:
Okay. Thank you, Mr. Angeleri. Mr. Lewis?

[00:12:25] Speaker 00:
May it please the court, my name is Damon Lewis for Appellee Mercedes-Benz.

[00:12:30] Speaker 00:
The board got this case entirely correct after all the proceedings and the expert testimony and a set of consolidated hearings have found that all the challenge claims of the 551, 029, and 503 patents are unpatentable based on Alden alone or in combination of Kobayashi or Harbers. These final decisions were correct in their respect and were supported by substantial evidence, which is the standard of view in this matter.

[00:12:53] Speaker 00:
I don't have time to go through all the issues.

[00:12:55] Speaker 02:
Start with classifying.

[00:12:57] Speaker 00:
Sure. That's the one I was going to go to right now, just first.

[00:13:03] Speaker 00:
Now, my brother-in-law on the other side says that the board did not apply the correct... Married to his sister?

[00:13:11] Speaker 04:
I'm just curious. Brother-in-law.

[00:13:15] Speaker 00:
Well, we are brothers at the bench, and this is law, so...

[00:13:23] Speaker 00:
I want to directly answer the questions that Judge Moore and Judge Cunningham are getting at, whether the claim requires multiple classes of objects to classify. The answer is no. The claim itself, Claim 48 in this instance, does not require any individual classification of multiple items. The only thing that the claim actually requires is that it has a first light determined based at least in part on classifying the object based on at least in part on the data, and then using that information, it then decides whether or not to dim the segment or not dim it.

[00:14:00] Speaker 00:
Now, there was a bit of discussion about some examples in the specification As your honors are probably aware, the specification is quite broad and includes things that have nothing to do with automotive applications such as headlamps. The specification also includes things such as highlighting different food items and utensils that have nothing to do with an automotive application.

[00:14:27] Speaker 03:
What would you contend is the best intrinsic support for the board's construction of classifying?

[00:14:33] Speaker 00:
The best, well, the board didn't construe it, left it as plain and ordinary meaning.

[00:14:37] Speaker 00:
And the good reason to do that is because none of the claim limitations require any extraneous limitations that the SPARO is trying to add via their dictionary definitions or the specifications examples.

[00:14:53] Speaker 03:
The board said plain and ordinary meaning, which is blah, blah, blah. I might be paraphrasing, but the board put some additional words beyond just saying plain and ordinary meaning, right?

[00:15:03] Speaker 03:
Do I misremember that with all of these different ones going on?

[00:15:06] Speaker 00:
I might be misremembering. I don't remember that. So I'm not going to say that you're misremembering.

[00:15:11] Speaker 00:
But that may be correct.

[00:15:14] Speaker 00:
But the long and short of it is that the claim language does not require the extra requirements that Sparrow was advocating, which would be to distinguish among multiple categories. The claim language itself doesn't require any kind of distinction between multiple categories. All it requires is that the censors...

[00:15:42] Speaker 00:
for lack of a better word, detect an oncoming vehicle and dim the segments that are facing that oncoming vehicle. And because the PTAB's charge is to only construe where it is necessary to solve the dispute between the parties, keeping it without the extra limitations is appropriate in that sense.

[00:16:03] Speaker 00:
Did I answer your question, Your Honor? Okay.

[00:16:05] Speaker 03:
I think so, but if I have another question, I'll just ask it, so don't worry.

[00:16:08] Speaker 00:
Okay.

[00:16:09] Speaker 03:
What about this issue with respect to the argument that there was maybe some conflict between another IPR in terms of this harbor's reference and the sensing fog or snow conditions?

[00:16:24] Speaker 00:
Right. You're looking now at the 1470 appeal, the last issue involving harbors.

[00:16:32] Speaker 00:
Now, it is true that Mercedes-Benz and VW both used Harbors, and if I'm not mistaken, that's the only overlap in terms of the references that the two parties used.

[00:16:45] Speaker 00:
The clarifying issue, I think, is that for both Mercedes and for VW, Harbors was a secondary reference. The primary references that were used were completely different. And so if the question is, why would you get one result when your name is VW, then you would get when your name is Mercedes. And the answer is with different inputs, you should expect or at least allow for the possibility of different outputs. Thank you.

[00:17:14] Speaker 02:
There's some requirement or some responsibility for the board. I think here is the same APJ that reached both conclusions, right? To drop a footnote or something to explain, whichever came second, to explain the discrepancy. I mean, we can surmise, I'm sure you better than I, what the differences are between the cases, but the board didn't tell us anything about that.

[00:17:38] Speaker 00:
That's right. I don't know if the circuit should apply a rule for the board below to differentiate the cases that come before, particularly when there's I think there are something like 13 IPRs on these three patents.

[00:17:57] Speaker 00:
But surmise it to say for purposes of your view at the federal circuit level is that substantial evidence supports the board's finding in our situation where substantial evidence may not have supported the board's finding in Volkswagen.

[00:18:13] Speaker 02:
Your reasoning in that regard is you're saying, yeah, the second reference, Hobbers, was the same, but the Alden here makes the difference?

[00:18:21] Speaker 00:
That's not quite my argument, Your Honor. And forgive me, I'm not as familiar with the Volkswagen record as I am with Mercedes' record. But in Mercedes' case...

[00:18:34] Speaker 00:
We presented expert testimony that went straight to the issue that Sparrow is now saying we didn't present below. So they're saying that, well, they want the ruling that they received in the VW case saying that Harvest does not disclose a sensor.

[00:18:54] Speaker 00:
In our situation, our petition and our expert testimony from the start at least made mention of the fact that HARBORS teaches the presence of a sensor. It says that when fog and snow is sensed, that we're going to shift the light towards the yellow end of the spectrum so as to provide a better visibility for the driver in such conditions.

[00:19:18] Speaker 00:
We actually had presented...

[00:19:21] Speaker 00:
evidence from Harbors on page 2 of the Harbors reference, which you can find on Appendix 821.

[00:19:32] Speaker 00:
Harber's actually goes into some detail about when the spectral output of the generated beam would be changed. It says that the autoelectric elements can be switched on and off with the spatial distribution of generated light beams depending on the conditions. It continues in that paragraph to note that the external conditions can include weather conditions, bright weather, fog, rain, snow, etc., and also includes that there may be some internal conditions such as the shape of the light beam which would be set or desired by the driver.

[00:20:06] Speaker 00:
And so the harbor's preference itself discusses how it's going to be aware of the weather conditions amongst other inputs, other stimuli, to alter the light output.

[00:20:26] Speaker 00:
That, in our opinion, and what we believe a person having ordinary skill in AR would understand, is teaching a sensor. Now, we haven't discussed it directly, but I believe that this also leads into one of the issues that we raised in our briefing about forfeiture.

[00:20:43] Speaker 00:
It is Mercedes' position that we did provide evidence, substantial evidence that the board could rely on to find that all of the limitations of the claims 29 and 69 of the 503 patent were met by the combination of alders with harbors.

[00:21:00] Speaker 00:
Had Sparrow raised at the board below, we don't think that Mercedes has presented enough evidence on this particular topic.

[00:21:10] Speaker 00:
We could have pointed the board back to where we had.

[00:21:14] Speaker 00:
And the record would have been more fulsome in that regard, but instead they chose to attack the Alden versus Harbors combination with completely different argument.

[00:21:26] Speaker 00:
And they left the record that way until it got here where they raised it for the first time. That is a violation of several Federal Circuit cases that say you can't come to the Federal Circuit with new arguments. You need to present the ones you presented below.

[00:21:42] Speaker 00:
So if they had done so, and this is the reason why the forfeiture doctrine exists, if they had done so, if they had presented this argument below, we could have pointed them directly to why we thought that harbors taught the sensed and sensed requirements as is required by claims 29 and 69.

[00:22:02] Speaker 03:
So because you asked about the classifying, even though I should ask the questions as opposed to you asking me, right, would be appendix page 516L. That's where it said classifying, and this would be in Appeal No. 2025-1306.

[00:22:16] Speaker 00:
Can you give me the page number you just left in, please?

[00:22:21] Speaker 03:
Yes, Appendix 5160.

[00:22:24] Speaker 03:
So at the top of 5160 is where it talks about the parties not appearing to dispute that classifying the object includes arranging or grouping the object according to some system or principle.

[00:22:36] Speaker 03:
So I do feel like the board pit forth the construction, but then goes on to say you don't need to further construe these claims for purposes of this decision.

[00:22:46] Speaker 00:
I'm sorry. I didn't catch the question. I apologize.

[00:22:56] Speaker 03:
I probably was really answering your question.

[00:22:58] Speaker 00:
Oh, okay.

[00:22:59] Speaker 03:
But I was pointing out how Appendix Page 5160 is where I saw the construction of classifying the object. We were talking about whether or not Mr. Irvin

[00:23:06] Speaker 00:
I understand.

[00:23:07] Speaker 03:
I don't think I misremembered. I think I remembered it.

[00:23:09] Speaker 00:
Yes, I understand. Yes, that's correct. The board below did understand that there wasn't a dispute, that these were the things that were included in the classification. And because there was a dispute, there was no need for the board to construe the claims further.

[00:23:21] Speaker 02:
Yeah, but his answer here, I don't know whether I agree with it, but I understand what he's saying. that the definition he put forth included to arrange or group in classes, and that that was not present in the board's construction. And I thought that was his gripe.

[00:23:40] Speaker 02:
Am I wrong? Is that the way you understood his argument?

[00:23:45] Speaker 00:
I understand Sparrow's argument, at least today, to be saying that the board should have taken the construction. I've lost my page. I apologize.

[00:23:58] Speaker 00:
the board should have used the construction distinguishing amongst multiple categories. And there may be some difference between what they said in their opening brief and what they said in their reply. I'm not certain. But the long and the short of it is that regardless of whether the construction was applied the way, regardless of whether the appellant got the construction they wanted, the board was presented and relied on substantial evidence that the Alden system does actually classify the oncoming vehicle differently from something else.

[00:24:43] Speaker 00:
If we stop for a moment.

[00:24:48] Speaker 00:
If we look at what Alden actually does, it detects spots that are above a certain threshold, and it defines these spots as two light sources that are near each other, and it uses triangulation to determine the distance, and it analyzes the intensity, size, and location of these dots, and it uses that information to determine which headlight sectors to dim. It is identifying and is classifying the presence of an oncoming vehicle to the exclusion of, for instance, the reflectors that Alton also discusses.

[00:25:19] Speaker 00:
And so even if we were to accept the additional limitations that Sparrow was advocating for, under the Plano meeting and a meeting that Sparrow is advocating, there is still a classification going on in Alden, and Alden does teach this limitation.

[00:25:34] Speaker 05:
Okay. Thank you very much. Thank you, Larry. We have some rebuttal time.

[00:25:55] Speaker 01:
Just a couple of points, Your Honor. With respect to harbors, I would just offer the Court, when we briefed it, we didn't have the A sites. We were citing the VW appeal.

[00:26:08] Speaker 01:
And the appendix hadn't been constructed yet. So I can give you the appendix sites to the VW findings that we cited on the Mercedes site. And they are at 74. So this is in the 1064 appendix, appeal number 2025-1064.

[00:26:27] Speaker 01:
pages 7479 and 7480 are the pages of the VW decisions that articulate the standard for what a petitioner needs to do to prove that Harbors teaches sensing fog and snow and find that Harbors does not sense fog and snow. So I just offer that for the court.

[00:26:48] Speaker 01:
I was trying to articulate you had asked me questions about the specification with respect to classifying and I was trying to articulate Specification examples that distinguish among multiple classes but use the term identified, those are in the board's decision at Appendix 5159. So those would be additional specification references that we believe support classification definition. And then the last thing, Your Honor, you had asked about the plain and ordinary meaning, or maybe I commented on it, but I just wanted to give you appendix sites.

[00:27:20] Speaker 01:
The dictionary definition is at Appendix 5157.

[00:27:26] Speaker 01:
The board's definition, which you asked counsel about, is at 5160. And what I was trying to communicate, I'm not sure if I did it effectively, is that the board's definition basically started from the dictionary definition but took out the word classes. So I'm not sure if I communicated it, but those are the appendix sites. Thank you, Your Honor. Okay.

[00:27:43] Speaker 05:
I think all counsel in this case has taken orders.