Cellular A 1-Homology from Bruhat Boundary Matrices of Split Semisimple Flag Varieties
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Mathematics. The day in mathematics: new theorems, new proofs, and what they open.
Ada: I'm Ada.
Emmy: And I'm Emmy, here with Kolya.
Ada: We're looking at the paper: "Cellular A 1-Homology from Bruhat Boundary Matrices of Split Semisimple Flag Varieties".
Emmy: Here's the gist of the paper. Let k be a perfect field of characteristic different from 2, we compute the cellular A1-homology of the flag varieties G/PΘ attached to split semisimple simply connected groups over k and describe the differentials in the cellular A1-chain complex concretely The purpose of this paper is to lift these integral boundary-matrix calculations to cellular A1-homology over a perfect base field k with char k!= 2.
Ada: Core Methodology The paper develops a method to compute the cellular A1-homology of split semisimple flag varieties by lifting integral boundary-matrix calculations to the motivic setting using Morel–Sawant’s cellular construction The chain groups are built from strictly A1-invariant sheaves, with Milnor–Witt Ktheory replacing the coefficient group Z of ordinary cellular homology.
Emmy: The oriented cellular A1-differential is given by ∂A1i = ηEΘi for i >= 2, and ∂A11 = 0.
Ada: This differential is related to the ordinary Bruhat boundary through the formula µ(v, w) = I degA1 Φ−1πΘ(v)ΦπΘ(w) (ht(γ∨) + 1)ϵη.
Title and authors: Ada: So we're looking at this paper, "Cellular A one-Homology from Bruhat Boundary Matrices of Split Semisimple Flag Varieties <ref:2607.02985#pg1,Cellular A 1-Homology from Bruhat Boundary Matrices of Split Semisimple Flag>." This work is about calculating the cellular A one homology for flag varieties attached to split semisimple simply connected groups over a perfect field k with characteristic different from two. It seems like they are taking some integral boundary matrix calculations and lifting them into this motivic setting using Morel Sawant's construction.
Emmy: Exactly, Ada. The key idea here is using Milnor Witt Ktheory instead of ordinary integers for the coefficient group in their cellular chain groups, which is what makes this a different flavor of homology than what we usually compute over a field like R or C. It's about getting an algebraic analogue of cellular homology that works nicely over these fields.
Kolya: From my side, I'm interested in how they handle the differentials. They define the oriented cellular A one differential as being equal to eta E Theta i for i greater than two, and zero for the first differential, which is a pretty standard setup but applied here with these new coefficients.
Ada: Right, and that specific formula for the differential is where things get really concrete. The paper says they relate this to the ordinary Bruhat boundary through a specific formula involving a quantity called mu of v w, which involves the height of some root and an eta factor. That seems like a direct bridge from classical combinatorics to this new setting.
Emmy: It is that comparison that's fascinating for me because it shows how the structure on the geometric side relates directly to these algebraic structures in Milnor Witt Ktheory. It’s not just a calculation; it’s establishing an equivalence between two worlds of mathematics.
Ada: So, what they apply this construction to is pretty broad. They use the type A coefficient formula for type A, and they extend it to the classical types B n, C n, D n for up to seven dimensions. Plus they cover the exceptional types F four through E six and E seven in their analysis of this paper.
Title and authors: Kolya: I wonder how robust this is when we move away from real fields R. The paper mentions that over k equals R, they recover the results of real flag manifolds, which is a good sanity check for the construction before going to more abstract perfect fields.
Emmy: That recovery over the real numbers is important because it validates that their method respects the known topological structures we see in these classical settings. It shows that there's a deep structural reason why this algebraic machinery works for those specific geometries.
Ada: And they do provide a detailed computation for SL three by B, which is a concrete example of this application before moving on to the full split flag variety of type F four. That kind of explicit calculation gives us something tangible to look at.
Kolya: I'm curious about what that SL three by B example actually yields in terms of homology groups. The paper claims that for SL three by B, there's only one non-zero cellular A one boundary, which is the codimension two boundary from length one cells to length two cells, and the result is zero minus eta minus eta zero.
Emmy: That specific result gives us a hint about the structure of the homology itself. It tells us that even in this motivic setting, there's a way for certain torsion elements to appear or disappear depending on the geometry of those boundaries.
Ada: And then they summarize it for type F four by saying that the resulting homology is given by H cell i X equals KMW i beta i plus M q greater than or equal to one KMW i over q eta m i q plus M q greater than or equal to one eta KMW i m minus one.
Kolya: That formula tells us a lot about the structure of the resulting homology sheaf. It shows that instead of just having integers and some simple torsion, you get these contributions from both KMW and its modification with eta factors.
Emmy: And what this implies is that the two-torsion we usually see in real flag manifolds doesn't just vanish when you move to this motivic setting <ref:2607.02985#pg1>. Instead, it gets split into two distinct sheaf-theoretic contributions, KM i and eta KMW i.
Title and authors: Ada: So to wrap up, the main implication is that we can lift those integral boundary matrix calculations—which are usually hard to handle abstractly—to cellular A one homology over any perfect field k with characteristic different from two. This gives us a finite computation for the homology in a range defined by definition two point one <ref:2607.02985#pg1>.
Kolya: If this method works for all these types up to seven dimensions, it suggests that we have a systematic way to handle the structure of these flag varieties using these motivic tools, even when we're dealing with more complex geometries like F four or E six.
Emmy: And the computational process they outline is quite detailed. It involves fixing Bourbaki numbering and normal forms, then enumerating Bruhat covers and computing boundary data like deleted positions and coordinate changes before forming those reindexed matrices E Theta i.
Ada: And the final step is performing a Smith decomposition of these reindexed matrices over integers to plug the elementary divisors into Theorem five point one, which gives us that homology structure we discussed earlier. It's a very systematic algorithm for getting from a geometric boundary to an algebraic homology group.
Kolya: I think the real power here is that they show every available matrix D Theta i has this cellular A one homology lift over k, which means the method is universally applicable for these cases within their defined scope.
Emmy: It's a very constructive result because it doesn't just give us a formula; it gives us the machinery to compute it explicitly by following those six steps they lay out. It’s about building an algorithm from the ground up.
Ada: So, in short, this paper on Cellular A one Homology from Bruhat Boundary Matrices of Split Semisimple Flag Varieties shows that integral boundary matrix calculations can be lifted into the motivic setting to give us explicit cellular A one homology sheaves over perfect fields k with characteristic different from two.
Kolya: It’s a solid result for the range they cover, and it gives us concrete tools to analyze the structure of these flag varieties using KMW theory, even when we're dealing with more complicated root systems.
Emmy: It certainly provides a strong link between the combinatorial structure of Bruhat boundaries and the algebraic invariants coming out of Milnor Witt Ktheory.
Ada: And that connection is what makes this work so interesting for us in number theory and geometry—it shows a consistent way to bridge those worlds.
The paper's summary: Ada: So, we've seen how they take those integral boundary matrix calculations and lift them into this motivic setting using Morel Sawant’s construction to compute cellular A one homology for flag varieties over perfect fields with characteristic different from two.
Emmy: Exactly, Ada. The big idea here is that they're replacing the usual integer coefficients with something called Milnor Witt Ktheory, which lets them study the structure of these flag varieties in a way that’s more algebraic than just pure topology.
Kolya: From my side, I'm looking at how they tie this to real numbers. They say that when you look at the split real flag variety over the reals, their motivic result simplifies to something we already know about integral homology of real flag manifolds.
Ada: Right, Kolya. And that simplification is key because it confirms that their machinery is consistent with what we expect from classical geometry before they move on to more abstract perfect fields.
Emmy: What really stands out to me is how they connect the oriented cellular differential directly back to the Bruhat boundary formula using a specific quantity called mu of v w, which involves the height of some root and an eta factor.
Kolya: That connection between the geometric boundary data and this algebraic formula for mu seems like a very direct way to see what's happening underneath.
Ada: And then they use that information to set up these reindexed matrices E Theta i, which are then subjected to a Smith decomposition over integers, leading to their final homology structure.
Emmy: That final structure, the one involving KMW and its modifications with eta factors, shows that even in this motivic world, we still get those interesting two-torsion contributions that we see in real flag manifolds.
Kolya: So what this means for us is that we can compute these homology sheaves over any perfect field with characteristic different from two, which is a huge step because it gives us a concrete formula where before there was mostly just integral boundary data.
Ada: Precisely. It means every matrix they've calculated has a lift in this cellular A one homology, and they’ve shown how to extract the Betti numbers and two-torsion ranks for real flag manifolds using the Smith data of those matrices.
Emmy: It shows that the topological information from the boundary is encoded in these KMW sheaves, which is really interesting because it gives us a way to see how geometry translates into algebraic invariants in this setting.
Kolya: I wonder if this method can be extended beyond type A and classical types B n, C n, D n for smaller dimensions, or if there’s some limit to what these boundary data hypotheses can handle.
Ada: The paper covers those types up to seven dimensions, so they've shown it works for a wide range of classical groups. They also address the exceptional types F four through E six and E seven in their analysis.
Emmy: And the computational algorithm they lay out is quite systematic, starting from fixing normal forms and then enumerating all the Bruhat covers, which is necessary to get all those boundary data points.
Kolya: It’s a lot of bookkeeping—fixing numbering, checking overlaps between Bruhat covers—but it sounds like a solid procedure for generating these matrices E Theta i that you need for the Smith decomposition.
Ada: The real punchline is that this approach provides a finite computation of the cellular A one homology sheaves over k in the range covered by their definition, confirming that every matrix D Theta i has this cellular A one homology lift over k.
Emmy: It’s constructive because it doesn't just give you a final formula; it gives you the steps to compute it explicitly using KMW theory for these geometric objects.
Kolya: So, if we look ahead, I wonder if these methods can handle even larger groups or more complex structures where the boundary data becomes exponentially harder to enumerate.
Ada: That's a fair question for future work. For now, they’ve established this powerful lifting technique over perfect fields with characteristic different from two for the flag varieties they cover.
The paper's improvements: Ada: So, we're looking at how they suggest improving their method for computing cellular A one homology by suggesting ways to refine the input data and the output structure itself.
Emmy: Right, Ada. They aren't just stopping at showing you *a* homology; they’re pointing out exactly where the structure of those KMW sheaves comes from, and how you can get finer control over that.
Kolya: What I find interesting is their discussion about the complexity of enumerating all those Bruhat covers; they are talking about ways to simplify that enumeration process if you're dealing with very large flag varieties.
Ada: They touch on how the boundary data hypothesis needs to be met, suggesting that pinning the root datum and parabolic type in one of the sources of boundary data is a requirement for their main computations to work as intended.
Emmy: That’s structural; it means the whole machinery relies on this specific setup being present somewhere in your input data, otherwise, the analogy between the boundary and homology breaks down.
Kolya: From a probabilistic view, if you're looking at these bounds for real flag manifolds, they are discussing what happens when you look at the tail of those distributions; they want to know if the constants survive when you apply this to more complicated systems.
Ada: They touch on refining how they compute those boundary coefficients mu of v w by explicitly mentioning the height of the root and a factor that depends on whether you have an eta or not, which is a way to get more precise control over that quantity.
Emmy: It’s about making sure the algebraic calculation precisely mirrors the geometric structure, and if you can pin down those factors better, you get a clearer picture of how the homology behaves.
Kolya: So it's less about finding a single number and more about understanding the range of possible outcomes for those quantities under different conditions.
Ada: They are definitely pushing toward making this an algorithmic tool, showing that every available matrix D Theta i has this cellular A one homology lift over k, which is a strong statement on its applicability.
Emmy: The implication is that if we can get better at controlling the input data and the coefficient computation, we can get a more rigorous handle on the structure of these spaces in general.
Kolya: It suggests that even when dealing with things as complex as type F four or E six, there are still systematic ways to analyze their underlying homology without just relying on brute force counting.
Ada: The paper shows that this lift works for a wide range of cases up to seven dimensions, which is a pretty good scope for a foundational result in this area.
Emmy: It lays the groundwork for future work by defining exactly what kind of input conditions are necessary to make these integral boundary calculations translate cleanly into KMW sheaf computations.
Kolya: So the next step for this line of research would be to see if we can push the dimensionality or the complexity of the group structure even further.
Conclusion: Ada: So, to wrap up, this paper on "Cellular A one Homology from Bruhat Boundary Matrices of Split Semisimple Flag Varieties" shows we can lift integral boundary matrix calculations into a motivic setting over any perfect field with characteristic different from two.
Emmy: That means the structure of these flag varieties is encoded in KMW sheaves, and we have an explicit algorithmic way to compute those sheaves using the Smith formula.
Kolya: I just want to say that this confirms every matrix they calculated has a cellular A one homology lift over k, which is a solid result for the range they cover.
Ada: Exactly. It gives us a finite computation of these homology sheaves over k in the range defined by their definition, which is pretty useful for understanding the geometry algebraically.
Emmy: The big implication here is that it provides a constructive link between classical combinatorial objects and algebraic invariants in KMW theory, which is something I find really compelling.
Kolya: It’s solid because it works for all those types up to seven dimensions, so we know this method has broad applicability across different root systems.
Ada: And they’ve shown that the topological two-torsion doesn't just disappear when moving to this motivic answer; instead, it gets split into these two distinct sheaf contributions.
Emmy: That split is what really matters because it shows that the torsion we see in real flag manifolds has a specific structure that survives this shift to the algebraic setting.
Kolya: I think for probabilistic applications, knowing how those constants survive the application when you move from real fields to perfect fields is important for understanding tail behavior.
Ada: Indeed. And they've given us a systematic way to analyze these flag varieties using these motivic tools, moving away from just relying on integral boundary data.
Emmy: It’s a constructive result because it doesn't just give you a final formula; it gives you the steps to compute it explicitly by following those six algorithmic steps.
Kolya: So the next step for this line of research would be to see if we can push the dimensionality or the complexity of the group structure even further.
Haoyang Liu, Tianle Liu
math.AG
Submitted: 2026-07-03
Updated: 2026-11-03
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Importance score: 83/100
The gist: Let k be a perfect field of characteristic different from 2, we compute the cellular A1-homology of the flag varieties G/PΘ attached to split semisimple simply connected groups over k and describe…
Key concepts
- Cellular A1-Homology
- This is a specific type of homology theory computed using sheaves on algebraic varieties over a field. It uses 'A1-invariant sheaves' and Milnor–Witt Ktheory as coefficients, providing a way to compute topological invariants in the motivic setting.
- Morel–Sawant's Construction
- This is the method used to lift integral boundary matrix calculations into the motivic setting. It allows mathematicians to translate classical integral homology results from real flag manifolds into a more general framework applicable over fields with different characteristics.
- Bruhat Boundary Matrix Calculations
- These are integral matrices derived from the structure of Coxeter groups and Bruhat covers of flag varieties. The paper uses these matrices as input to compute the cellular A1-homology, linking combinatorial data to sheaf computations.
- Milnor–Witt Ktheory (KMW)
- This is a coefficient group used in the cellular A1-homology instead of standard integer coefficients. It plays a crucial role in describing the structure of the resulting homology groups, particularly how 2-torsion is distributed among different sheaf components.
Terminology
Summary
Let k be a perfect field of characteristic different from 2, we compute the cellular A1-homology of the flag varieties G/PΘ attached to split semisimple simply connected groups over k and describe the differentials in the cellular A1-chain complex concretely The purpose of this paper is to lift these integral boundary-matrix calculations to cellular A1-homology over a perfect base field k with char k! p. 0= 2.
Core Methodology
The paper develops a method to compute the cellular A1-homology of split semisimple flag varieties by lifting integral boundary-matrix calculations to the motivic setting using Morel–Sawant’s cellular construction The chain groups are built from strictly A1-invariant sheaves, with Milnor–Witt Ktheory replacing the coefficient group Z of ordinary cellular homology. The oriented cellular A1-differential is given by ∂A1i = ηEΘi for i >= 2, and ∂A11 = 0. p. 2 This differential is related to the ordinary Bruhat boundary through the formula µ(v, w) = I degA1 Φ−1πΘ(v)ΦπΘ(w) (ht(γ∨) + 1)ϵη. p. 2
Boundary Data and Input
The construction applies uniformly to the type A coefficient formula, to the type Bn, Cn, Dn for n ≤ 7, and to the exceptional types for F4, E6, E7. The required orientation data are precisely the deletion signs, degrees of coordinate changes, and coroot-height parities appearing in the signed Bruhat boundary formula. p. 1 The boundary-data hypothesis requires that the pinned root datum and parabolic type occur in one of the sources of Bruhat boundary data. p. 3
Homology Structure
The Smith decomposition of the reindexed matrices EΘi leads to a formula for cellular A1-homology Hcell i(XΘ) ∼= (KMW i)bi ⊕ Mq>=1 (KMW i /qη)mi(q) ⊕ Mq>=1 (qηKMW i)mi-1(q). In the cases where the associated split real flag variety XΘ,R has integral homology of the form Hj (XΘ,R(R), Z) ∼= Zβj ⊕ (Z/2)Tj, this simplifies to Hcell i(XΘ) ∼= (KMW i)βi ⊕ (KM i)Ti ⊕ (ηKMW i)Ti-1. p. 2
Applications and Results
The method allows for the lift of integral boundary computations to cellular A1-homology over any perfect field k with char k!= 2. For the full split flag variety of type F4, the resulting homology is given by Hcell i(X) ∼= (KMW i)βi ⊕ (KM i)Ti ⊕ (ηKMW i)Ti-1. p. 2 This result shows that the topological 2-torsion does not disappear in the motivic answer but is refined into two sheaf-theoretic contributions KM i and ηKMW i. p. 2
Algorithmic Summary
The computation involves a six-step process:
-
Fix Bourbaki numbering, a split pinning, horizontal normal forms for all w ∈ WΘ, and the compatible Morel–Sawant normal expressions for the complementary orientations. p. 3
-
Enumerate the length-graded set WΘ and all Bruhat covers w ⋗ w′. p. 4
-
For each ordinary cover a ⋗ b, compute the normal-form boundary data: the deleted position I(a, b), the root γ(a, b), and the degree of the coordinate change determined by commutations and braid moves. p. 2
-
Form DΘ i = (cΘ(w, w′)/2). p. 2 Check DΘ i-1DΘ i = 0. p. 4
-
Form the length-reversing permutation πΘ(w) = w0wwΘ and the matrices Pi[w] = [πΘ(w)]. p. 1 Then form the codimension-indexed matrices EΘ i = P−1i-1DΘ i Pi. p. 9
-
Compute a Smith decomposition of EΘ•, and insert the elementary divisors into Theorem 5. p. 01.
Example Computation
For the example SL3/B, the only non-zero cellular A1-boundary is the codimension-two boundary from the two length-one cells to the two length-two cells, resulting in ∂A12 = 0 −η −η 0. This leads to Hcell 1(SL3/B) ∼= (KM 1)2 and Hcell 2(SL3/B) ∼= (ηKMW 2)2. p. 2 For the full split flag of type F4, the homology is given by Hcell i(X) ∼= (KMW i)βi ⊕ (KM i)Ti ⊕ (ηKMW i)Ti-1. p. 2
Conclusion
The paper demonstrates that the cellular A1-homology of split semisimple flag varieties over a perfect field k with char k!= 2 can be computed by lifting integral boundary matrix calculations to the motivic setting. p. 1 This provides a finite computation of cellular A1-homology sheaves over k in the range covered by Definition 2. p. 31. The resulting homology is given by the Smith formula of Theorem 5. p. 01. This confirms that every available matrix DΘ i has a cellular A1-homology lift over k. p. 2
References
[1] A Bj¨orner and F. Brenti, Combinatorics of Coxeter Groups, Graduate Texts in Mathematics 231, Springer, 2005
[2] R. W. Carter, Simple Groups of Lie Type, Pure and Applied Mathematics 28, Wiley, 1972
[3] T. Hudson, A Matszangosz, and M Wendt, Chow–Witt rings and topology of flag varieties Journal of Topology 17 (2024), no. p. 1 4, e70004
[4] R. R. Kocherlakota, Integral homology of real flag manifolds and loop spaces of symmetric spaces Advances in Mathematics 110 (1995), 1–46
[5] J. Lambert and L. Rabelo, A correspondence between boundary coefficients of real flag manifolds and height of roots Journal of Lie Theory 32 (2022), 431–446
[6] J. Lambert and L. Rabelo, On the computation of homology of type A real flag manifolds arXiv:2306. p. 001127, 2024
[7] J. Lambert Silva and L. Rabelo, Integral Homology and Poincar´e Polynomials of classical and exceptional Real Flag Manifolds arXiv:2605. p. 004944, 2026
[8] J. Lambert, jlambert33/FlagManifoldTopology: v1.0.1, Zenodo, 2026 https://doi. p. 0org/10.5281/zenodo.19871728
[9] H Liu and K Peng, Cellular A1-Homology of Smooth Toric Varieties arXiv:2505.04520, 2025
[10] F Morel and A Sawant, Cellular A1-homology and the motivic version of Matsumoto’s theorem Advances in Mathematics 434 (2023), 109346
[11] F Morel, A1-Algebraic Topology over a Field Lecture Notes in Mathematics 2052, Springer, 2012
[13] L. Rabelo and L. A. B. San Martin, Cellular homology of real flag manifolds Indagationes Mathematicae 30 (2019), no. p. 0 5, 745–772
[14] University of California, Santa Barbara Email address: haoyangliu@ucsb.edu University of Southern California Email address: tianleli@usc.edu
--- Page 1 ---
CELLULAR A
1-HOMOLOGY FROM BRUHAT BOUNDARY MATRICES OF SPLIT SEMISIMPLE FLAG VARIETIES HAOYANG LIU AND TIANLE LIU Abstract. Let k be a perfect field of characteristic different from 2, we compute the cellular A1-homology of the flag varieties G/PΘ attached to split semisimple simply connected groups over k and describe the differentials in the cellular A1-chain complex concretely
--- Page 2 ---
CELLULAR A
1-HOMOLOGY FROM BRUHAT BOUNDARY MATRICES OF SPLIT SEMISIMPLE FLAG VARIETIES HAOYANG LIU AND TIANLE LIU The purpose of this paper is to lift these integral boundary-matrix calculations to cellular A1-homology over a perfect base field k with char k!
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Improved AI system can compute cellular A1-homology for split semisimple flag varieties over perfect fields with characteristic different from 2, using Morel-Sawant's construction.
-
Improved AI system can explicitly describe the differentials in the cellular A1 chain complex by computing the oriented cellular boundary differential ∂A1i = ηEThetai for various flag types, including type A and classical types Bn, Cn, Dn for n <= 7.
-
Improved AI system can perform a chain-level comparison between integral Bruhat boundary matrices and Milnor-Witt sheaf computations by calculating the motivic boundary coefficient µ(v, w) using the formula µ(v, w) = ⟨−1⟩I degA1 Φ−1πΘ(v)ΦwI (ht(γ∨)+1)ϵη.
-
Improved AI system can determine the Smith decomposition of the reindexed matrices EThetai and apply Theorem 5.1 to compute the cellular A1-homology sheaf Hcell i(XTheta), yielding Hcell i(XTheta) ∼= (KMW i)bi oplus M q>=1 (KMW i / qη) mi(q) oplus M q>=1 (qηKMW i) mi-1(q).
-
Improved AI system can compute the Betti numbers and 2-torsion ranks of real flag manifolds by using the Smith data of DTheta i, which is equivalent to computing Hcell i(XThetaR) ∼= Zβj oplus (Z/2)Tj.
Sources
- On the computation of homology of type A real flag manifolds
- Integral Homology and Poincar'e Polynomials of classical and exceptional Real Flag Manifolds
- Cellular $\mathbb{A}^1$-Homology of Smooth Toric Varieties
Related papers
- K-unstable Toric Varieties and Secondary Polytopes
- The commuting scheme of g 2 and a degree formula for commuting schemes
- Stacky geometry and logarithmic topology of transversely affine foliations
- Monodromy and vanishing cycles for complete intersection curves
- Characteristic cycles and Gevrey irregularity in hypergeometric families
- Tautological classes on KSBA spaces