[00:00:00] Speaker 04:
Before we get started, I just wanted to welcome and thank Judge Seabird for sitting with us today and yesterday.

[00:00:07] Speaker 04:
We always appreciate the help, and we enjoy having our district court judges and colleagues come and join us on the bench.

[00:00:14] Speaker 02:
Thank you very much. Great pleasure.

[00:00:17] Speaker 04:
The first case this morning is Idea Hub, Inc. versus Unified Patents, LLC, number 24-1684. Mr. Newman, when you're ready.

[00:00:29] Speaker 03:
May it please the court.

[00:00:31] Speaker 03:
This case is about whether the board can apply collateral estoppel based on a supposed identity of issues without first construing disputed claim terms.

[00:00:49] Speaker 03:
It turns fundamentally on whether the term for, as used in the 702 IPR, is identical to the term associated with that's at issue here. The Board never performed a proper claim construction on either of those terms, never resolved the party's dispute with respect to those terms, and never fully detailed why its decision with respect to those terms mattered, other than to say that there was a lack of substantial evidence proving that the terms were different.

[00:01:28] Speaker 01:
Just to clarify, it's my understanding that you're relying on the intrinsic record for an understanding or interpretation of what for means versus associated with means. Is that correct?

[00:01:42] Speaker 03:
That's right. Well, the board didn't perform that. We think that that should be performed. That claimed construction analysis should be performed by the board in the first instance, and they should look to the intrinsic record in the first instance.

[00:01:56] Speaker 01:
You don't think that the board interpreted the terms as having the same meaning?

[00:02:01] Speaker 03:
The board found that there was insubstantial evidence to find that the meanings were different, which is sort of a different situation. This court has long held that different claim terms are to be afforded different claim scopes. And that's a presumption. It could be rebutted, but that presumption applies. Here, the board did exactly the opposite. It presumed that the terms were the same unless patent owner proved that they were different, which is two errors. First, failing to provide the presumption, but second, putting the burden on patent owner to prove that there's not an identity of issues with respect to collateral estoppel.

[00:02:39] Speaker 03:
that was also legal error.

[00:02:41] Speaker 04:
It's a much different analysis to say that... Do you agree with me to the part in the board's decision where you said they switched the burden to you?

[00:02:51] Speaker 03:
Sure.

[00:02:52] Speaker 04:
I think they're just describing the outcome, which is these differences are insubstantial.

[00:02:59] Speaker 03:
Right, so it's on Appendix 24 where there's a single paragraph in which...

[00:03:05] Speaker 03:
the Board makes that conclusion, and you're exactly right, Your Honor. It's just a conclusion that the Board comes to without setting forth a reasonable analysis of why the construction, why they adopted that understanding, and also why their understanding actually rendered the issues substantially identical with respect to the collateral benefits

[00:03:35] Speaker 04:
I think you must have misunderstood my question. Where do you think they switched the burden to you to prove a lack of identity in order to avoid collateral stopover? I mean, we're looking at two different terms, two different phrases that are pretty close. One is determining a prediction mode associated with the current block, and one is determining a mode for a neighboring block.

[00:04:02] Speaker 04:
Those are the two different claims we're looking at to determine whether there's collateral stop-off.

[00:04:07] Speaker 04:
They have mostly the same words. There's a few extra words in one versus the other that don't seem to make much difference. The only difference you seem to key on is for or associated with.

[00:04:18] Speaker 04:
You're looking at trying to convince us that for and associated in the abstract means something different, but for and associated are complex technical terms. So what we're really looking at is the whole phrase and whether it's different or not.

[00:04:33] Speaker 03:
That's exactly right, Your Honor. The phrase being determining an intermode for a neighboring block, which would imply that the neighboring block's intermode, the one that was used to decode the neighboring block, is what needs to be determined. And that's what the 702 IPR decided, and that's behind us. When you're talking about determining an intermode, associated with a neighboring block. That's a much broader situation. Now we can look at pixel pairs to determine the directionality.

[00:05:05] Speaker 04:
The directionality of the... I don't understand that at all. What's the difference between the intermode for a neighboring block and associated with a neighboring block? They're both talking about a mode, an intermode, and for the block. They're not talking about... Neither of these say pixels. They say associated with the block or the block.

[00:05:27] Speaker 03:
For the block, that's right.

[00:05:28] Speaker 04:
So can you point to me anything in the specification or anywhere that suggests what you're trying to do is have some difference between associated and for?

[00:05:40] Speaker 03:
Right, so there's three places that I would look. In the specification, the board actually pointed to a part of the specification in which they said an image associated with the neighboring blocks, which I think is important because an image associated, there's one image with thousands of blocks that comprise that image. which is different than having an image that's for the block, inside of the block. That shows the difference between associated with and how much broader it is than for.

[00:06:11] Speaker 03:
But also in the prostitution history.

[00:06:13] Speaker 01:
I don't understand that. I just wanted to ask, I mean, going back to your initial point, was that you said that the board did not construe these terms. But what you're referring to is the discussion on page A22 of the appendix where the board, although I agree it's a little cursory, I think construed the terms and including the reference to the specification that you're discussing right now. And what they said is there's nothing to support the conclusion that associated with has a broader scope than four in the context of the claim language, the specification, and the prosecution history.

[00:06:51] Speaker 01:
So I'm having a hard time understanding why you're saying there was no claim construction here. Yes.

[00:06:56] Speaker 03:
Thank you for that question, Judge Stoll. That's exactly the place where the board goes wrong, right? Their presumption is that there's no difference here. It's not a presumption.

[00:07:08] Speaker 04:
It's just their finding. Is there legal conclusion that these claims have no legally significant difference? That's claim construction. They're allowed to do that.

[00:07:18] Speaker 03:
But they say there's no substantial evidence that they're not, as opposed to finding that the… They don't say that.

[00:07:24] Speaker 01:
Looking at the paragraph on page A22, there's nothing about substantial evidence or anything like that. They're looking at the intrinsic evidence, and I guess there's intrinsic evidence, too, in the form of dictionary definitions, and say that, you know, it sounds like a legal conclusion. I understand that page A24, in talking at overall about the identicality of the issues for collateral stopover purposes, uses the phrase insufficient evidence of record.

[00:07:55] Speaker 01:
But I'm asking you specifically about the discussion on page A22, which is two pages prior to that.

[00:08:02] Speaker 01:
And you've taken the position today that there was no claim construction done. So that's what I'm questioning right now.

[00:08:07] Speaker 03:
But I don't think that the Board performed a claim construction under Phillips as this Court has described and told the Board what to do and how to perform that construction. Instead, it just comes up with its conclusion that these are the same.

[00:08:24] Speaker 02:
In a related question, I'm not sure I understand. What more do you think they should have done? I understand you disagree with the board's decision, but where is the absence of a reasonable explanation? They give you an explanation. You just don't like it. But how is it inadequate?

[00:08:41] Speaker 03:
Well, it's inadequate, Your Honor, because it doesn't address the dispute that the parties had below with respect to what these terms meant. It doesn't – so the parties, one party – Well, they said it means the same thing.

[00:08:52] Speaker 04:
You said it means different things, and the board said they agree with them. It means the same thing. That's the dispute.

[00:08:59] Speaker 03:
Well, in the first instance, the board in the institution decision said that the term required the determination of an intro mode for or – used to decode the neighboring block. And then in the final decision, it backed away from that and it didn't tell us why. I don't know whether the board still requires that it needs to be the intermode used to decode the neighboring block or not.

[00:09:26] Speaker 03:
The board had an obligation to tell us whether or not that is what it interpreted this claim to be.

[00:09:31] Speaker 01:
Isn't a reasonable inference to say if they had the view that four means the same thing as associated with and lighted the intrinsic and extrinsic evidence that that is what they're saying?

[00:09:45] Speaker 03:
I'm not sure. I'm not sure what they're saying. It's not that hard.

[00:09:50] Speaker 04:
They found collateral estoppel. So they must have concluded that four and associated mean the same thing. Otherwise, they wouldn't have found collateral estoppel.

[00:09:59] Speaker 03:
Well, that might be a reasonable inference, Your Honor, but under Chainery, we're not allowed to infer what the Board did, especially with... I don't think we have to do much inferring.

[00:10:09] Speaker 04:
They said it doesn't matter.

[00:10:11] Speaker 04:
To the extent it matters, right? I mean, the stuff you just pointed me to, you determined it's not necessary to expressly construe the claims because it doesn't make any difference.

[00:10:24] Speaker 03:
Well...

[00:10:26] Speaker 03:
I would disagree with that.

[00:10:27] Speaker 04:
Your Honor, I think that is... I know you disagree with their conclusion, but that's their work.

[00:10:30] Speaker 03:
I don't only disagree with their conclusion. I also disagree with their methodology. They should not just say, we're not going to construe this. They're the same.

[00:10:38] Speaker 04:
They don't have to construe stuff if it doesn't have any impact on their ultimate conclusion.

[00:10:45] Speaker 04:
And if their view is, these claims are similar enough, or are the same, we're not going to give you an official construction. They mean the same thing, so collateral stoppage.

[00:10:57] Speaker 04:
You read that reasoning in the opinion, right?

[00:11:00] Speaker 03:
I think that's what they did. I think they came up with a conclusion without adequately describing why they came to that conclusion and why it matters.

[00:11:09] Speaker 04:
Well, the conclusion is these two claim terms are substantially similar, so collateral estoppel applies, and we needn't construe anything because it would make no difference.

[00:11:20] Speaker 03:
But in order to determine whether two claim terms are substantially identical, first of all, you need to overcome... They just told you it was substantially identical when they told you that collateral stop will apply. But they need to overcome the presumption that the claim terms have different scopes.

[00:11:36] Speaker 01:
So let me ask you something. Okay, so there's a presumption that associated with and for have different meaning because your view is that when different terms are used in different claims, even though they're different patterns, that there's a presumption... That's a mild presumption, right, that the words have different meanings. But it doesn't stop there, right? You have to look at the context of the claim language, and you have to also look at the specification and the prosecution history.

[00:12:04] Speaker 01:
And then what is your challenge to that to the extent I read the opinion as the board doing that and saying, nonetheless, I think these are the same in the context of that evidence. And so I think you've got an argument for why that's not so. So what's your best argument for why these words should be given different meanings in the context of the specification, for example?

[00:12:27] Speaker 03:
In the context of the specification, as I mentioned, the use of the image that's associated with the neighboring blocks, but also the prosecution history.

[00:12:35] Speaker 01:
But in the context of the claim language here, which isn't talking about the image, okay? The full claim language here is that it's either determining an intra-prediction mode associated with that neighboring block. or determining an intra-prediction mode for a neighboring block. Why is it that I should understand those in the context of the claims and in the context of this specification to be different?

[00:13:06] Speaker 03:
Because naturally they are. For is a possessive. It's the intra-mode that's in the block. It's the one that the block owns, as opposed to associated with can be Any derivation from the blocks, anything that's related to the blocks, anything that's connected to the blocks.

[00:13:22] Speaker 00:
What in the specification teaches the broader meaning?

[00:13:28] Speaker 03:
Many of the examples that have been read out of the claims in the 702 IPR.

[00:13:34] Speaker 01:
So your view is that when it says associated with, that includes directionality? Yes.

[00:13:38] Speaker 03:
Absolutely, and there's great evidence that the other side's expert actually agrees that directionality corresponds. That's getting into the weeds. This is something that the board should evaluate on the merits.

[00:13:53] Speaker 03:
I see I'm out of time, so...

[00:14:07] Speaker 04:
of your last name?

[00:14:09] Speaker 05:
That's Chen, Your Honor.

[00:14:10] Speaker 04:
Chen, okay.

[00:14:12] Speaker 04:
Oh, I have a handy little chart here that I didn't look at that would explain that to me. No problem. Thank you.

[00:14:17] Speaker 05:
Good morning, Your Honors. May it please the Court, Michael Chen for Unified Patents. The Board correctly precluded IdeaHub from relitigating the written description issues that this Court and the Board already decided in the prior IPR.

[00:14:31] Speaker 01:
Why do you think, what do you think the Board did with claim construction on associated with and for?

[00:14:36] Speaker 05:
I think the board absolutely construed the issue that was dispositive here, which is that those two phrases mean the same thing in the context of these claims.

[00:14:46] Speaker 01:
Why do you think associated with, which is in plain meaning could be broader than for, why do you think? that it has this exact same meaning as for in this case?

[00:14:57] Speaker 05:
Because I think the context matters. These are prepositional phrases that are linking the neighboring intra-prediction mode with the neighboring block. So it's describing the connection between a parameter and the block that the parameter goes with. That's the same connection whether you use the preposition for... What is your response to the notion that different terms have different meanings? So I think that...

[00:15:22] Speaker 05:
That presumption, to whatever degree it carries any force, is weakened in this context, whereas Judge Stoll pointed out these are not terms used in the same patent. They're not in the same claim. They're not even in the same patent. These are across patents. And I think the rest of the context shows that the meaning is actually the same. If you look at the specification, I have a hard time understanding why the other side says that the image associated with the neighboring block, that phrase in the specification supports them.

[00:15:53] Speaker 05:
I think it supports us because that is clearly talking about the image that is on the neighboring block, possessed by the neighboring block, whatever the other side says for means. And so the specification is using the phrase associated with in the same way that the other side says for means.

[00:16:09] Speaker 01:
Can you show us why that's so? I mean, could you look at the language of the specification that the board relied on? And is there something in there to demonstrate conclusively that it's talking about the image for the block as opposed to some other image? Yeah, I think you can look...

[00:16:31] Speaker 05:
So the image associated with, that's, so we're looking at column six of the patent. So that's line 43, image associated with the neighboring blocks. If you skip down to line 50, it talks about this image, which is reconstructed for the neighboring blocks. So that is literally using those phrases interchangeably. But I don't think the court actually needs to go anywhere near this issue.

[00:17:01] Speaker 05:
What the board held was that these two phrases in context mean the same thing, looking at both the specification and, frankly, the lack of a meaningful difference in the dictionary definitions. But I think you could completely set... this for versus associated with issue aside on the merits, and affirm on a completely independent ground, which is that IdeaHub faces a separate fatal written description problem, which is that the specification does not provide written description support for determining a singular problem.

[00:17:32] Speaker 05:
intra-mode associated with the neighboring block, that issue carries over exactly from the prior IPR, and it is unaffected by any difference, even if there were one, between for and associated with. And so that's an independent reason why there's no material difference between the claim here and the one in the prior IPR, so collateral estoppel applies, and it's an independent reason why they haven't established the prejudice necessary to reverse the board on appeal. Okay.

[00:18:00] Speaker 05:
I'm happy to answer more questions about for versus associated with on the merits, but I do think that this independent ground is another way and maybe a simpler way to affirm.

[00:18:10] Speaker 02:
Going back to the other issue, do you think it would have been preferable for the board to have been more explicit about whether or not they're construing a claim or not? Do you think they should have done that?

[00:18:24] Speaker 05:
Sure, Your Honor. So this gets to the procedural challenge that they're raising, which in my mind is a little bit separate from the merits of whether forward versus associated with are actually the same. I think on the procedural challenge, it's even clearer that the board satisfied its obligations. It clearly did address the scope. of those two claims, you can look at not just the paragraph that's already been discussed this morning on Appendix 22, but if you go to Appendix 24, the board says that it rejects the conclusion that, and then it quotes the language from this patent here, and then it says it does not differ in scope from, and then it quotes the language from the claim in the IPR.

[00:19:04] Speaker 05:
So it's clearly construing the scope of the language in the claims, and it's concluding that there is no difference in scope between associated with and for. So as to the procedural challenge, I think it's absolutely clear that the board satisfied its obligation to reach the holding that was necessary to decide the case, and also to provide an explanation, which it did on Appendix 22. But again, the board also, in addition to resolving that for versus associated with, it also resolved this other independent ground that I mentioned. And that is the carryover between Appendix 22 and Appendix 23.

[00:19:38] Speaker 05:
It also reached this other issue of whether the claim requires determining a singular interim mode. And again, that is independent of any gap between for and associated with. You can look at the board's reasoning on this. You can look at the party's arguments on this. None of it has anything to do with any delta between for and associated with. Either way, the operative term determining requires determining a singular intro mode. And either way, the specification, which is the same here as in the prior IPR, does not provide written description support for that.

[00:20:15] Speaker 05:
I think there's one other loose end to tie up if you're going down that path, which is a new theory that IdeaHub raised for the first time in its reply brief, and although it wasn't discussed this morning, this is our only opportunity to respond to it, so I want to take a chance to do that.

[00:20:30] Speaker 05:
So in our read brief, we pointed out that IdeaHub had not identified any prejudice because they hadn't identified any way where their claim construction arguments would affect the written description conclusion. They came back in the reply brief with a brand new theory that the specification provides written description support for determining a singular intro mode when the preferred embodiment identifies a lowest cost mode to be used to decode the current block. So I think if the court got there, the court should reject that theory for a number of reasons.

[00:21:04] Speaker 05:
The first is forfeiture. It was raised for the first time in the reply brief. It wasn't in the opening brief, so it was forfeited on appeal. It was never presented to the board, so it was forfeited before the board. It's actually contrary to what IdeaHub said about the specification to the board. I think the clearest place to look for that is Appendix 408, where they say the thing that they are now saying is written description support for limitation 1.1. At Appendix 408, they said it actually goes to limitation 1.2. So I think it's absolutely forfeited, and that's enough to set aside that theory.

[00:21:36] Speaker 05:
I think the theory also fails in the merits. Their new theory in the reply brief is that the neighboring intra-prediction mode is the same as the current intra-prediction mode. That cannot possibly be. Those are different terms in the same claim of the patent. And actually, the patent requires intermediate steps to get from one to the other. That's limitation 1.2 and the wherein step starting at 1.8.0.

[00:21:59] Speaker 05:
So those two things cannot possibly be the same thing. So that's another reason to reject the new theory in the reply brief. And with that theory out of the way, there's no other argument for why the specification might provide written description support for determining a singular neighboring intro mode. And again, that's an independent reason to affirm without having to get into four verses associated with on the merits.

[00:22:24] Speaker 05:
If the Court has no further questions.

[00:22:26] Speaker 04:
Thank you, Mr. Chair.

[00:22:27] Speaker 05:
Thank you.

[00:22:31] Speaker 04:
Mr. Newman, I'll give you three minutes for a rebuttal. Thank you, Your Honor.

[00:22:36] Speaker 03:
So just to address the forfeiture issue, in our blue brief at 29, we absolutely say that under the associated with... It describes embodiments in which a single intermode is determined in connection with and is associated with evaluating the directionality of the neighboring block. So that's not forfeited. I want to just... let you know what we're asking. We think that the board should be required to do its job before you all have to affirm or deny.

[00:23:12] Speaker 03:
And so we're asking for the court to vacate the decision on collateral estoppel, which also necessarily vacates the decision on PGR eligibility and on written description because they rise and fall with collateral estoppel in this instance.

[00:23:28] Speaker 03:
and remand with an instruction to the court to perform a claim construction analysis consistent with this court's instruction in Phillips.

[00:23:41] Speaker 03:
And with a note that there is a presumption that needs to be overcome.

[00:23:47] Speaker 03:
I want to... points to a case, Parker Vision, that is very similar to this, where the lower court simply gave the conclusion that the two claim terms were substantially identical. And that was found to not be sufficient, that the court remanded and told the lower court, this court remanded and told the lower court, to perform its claim construction analysis properly before it...

[00:24:17] Speaker 03:
and to address everything on the merits after he did that. And that's all we're asking this court to do, to remand and tell the board to do its job so that we can be heard on the merits. And we think that when that happens, there was a lot of discussion recently about whether there's written description support. We don't know what the board thinks right now with respect to the written description, and that's the prejudice that IdeaHub is facing here. Because of the collateral estoppel, the board sidestepped actually having to engage on the merits as to the written description in this patent.

[00:24:54] Speaker 03:
And so we think that when this is remanded, if the board has to grapple with those issues, those issues, they will find in our favor. And we want to make sure that the board actually understands this technology, and it needs to prove to you all that it does and that its decision-making is well-founded and grounded in the facts.

[00:25:24] Speaker 04:
Thank you.

[00:25:24] Speaker 03:
Thank you.