--- title: 'ReMake: Structured Revision Across Modalities' url: https://www.emergentmind.com/topics/remake type: topic --- # ReMake: Structured Revision Across Modalities Searching arXiv for the specific ReMaKE paper and closely related "remake" usages to ground the entry in current arXiv records. In current arXiv usage, **“ReMake”** and cognate forms such as **“ReMaKE,” “Remake,”** and **“re-makeup”** do not denote a single method or benchmark. They instead name a family of operations that revise, normalize, reconstruct, or selectively re-curate an existing object rather than generating it ex nihilo. In large language models, ReMaKE is a retrieval-augmented framework for multilingual knowledge editing; in video generation, ReVideo formulates “remake a video” as local editing with joint content and motion control; in 3D face graphics, BareSkinNet treats de-makeup and de-lighting as the prerequisite for re-makeup; and in dataset construction, SEED produces the compact corpus **Honeybee-Remake-SEED-200K** through weighted independent set selection [2312.13040][2405.13865][2209.09029][2605.15691]. Other uses are broader but structurally related, including slide remaking in digital pathology, screen-remake attacks in face anti-spoofing, the “remake” of a mathematical characterization, and even a modeled electoral rematch [1803.07240][2311.17583][2008.07631][2112.13713]. This suggests that, across fields, “ReMake” functions as a constrained transformation paradigm: the target is edited, normalized, or re-instantiated while preserving selected structure from an antecedent object. ## 1. Research scope and recurring meanings The term appears in several technically distinct senses. In some papers it is an acronym, as in **Retrieval-augmented Multilingual Knowledge Editor (ReMaKE)**; in others it is a task label, as in **“Remake a Video”**; in others it is an ordinary noun or verb, as in **remake the slide**, **screen remake attacks**, **break and remake of contacts**, or **a 2022 remake** [2312.13040][2405.13865][1803.07240][2311.17583][1706.05898][2112.13713]. The shared research pattern is that an existing artifact is not discarded but reworked under explicit constraints. | Domain | Meaning of “ReMake” or “remake” | Representative source | |---|---|---| | Multilingual LLM editing | Retrieval-augmented prompt-based knowledge editing across languages | ReMaKE [2312.13040] | | Video generation | Local video editing with joint content and motion control | ReVideo [2405.13865] | | Face graphics | De-makeup and de-lighting as a prerequisite for re-makeup | BareSkinNet [2209.09029] | | Data curation | Task-targeted dataset re-curation | SEED / Honeybee-Remake-SEED-200K [2605.15691] | | Digital pathology | Remake or manually review low-quality slides | SlideNet [1803.07240] | | Face anti-spoofing | Screen replay or screen-remake attack category | CLIPC8 [2311.17583] | This range matters because the term is not merely rhetorical. In each case it identifies a specific intervention point in a workflow: post hoc factual correction in LLMs, localized regeneration in video, normalization before cosmetic transfer in 3D graphics, compact dataset reconstruction for instruction tuning, or quality-gated re-preparation in microscopy. The term therefore marks a shift from one-shot generation toward iterative, structure-preserving revision. ## 2. ReMaKE as multilingual knowledge editing In the most explicit acronymic usage, **ReMaKE** stands for **Retrieval-augmented Multilingual Knowledge Editor**, a model-agnostic method for updating LLM behavior when knowledge is edited in one language and queried in another [2312.13040]. The central problem is multilingual knowledge editing: if a counterfactual fact is supplied in language \(l_2\), the model should answer correctly when queried in language \(l_1\). The paper formalizes retrieval over a multilingual knowledge base \(K_{12}=\{k_1^{l_2},\dots,k_n^{l_2}\}\) by \[ k_{i^*}^{l_2}=R(x^{l_1})= \begin{cases} k_{i^*}^{l_2}, & \text{if } f_e(x^{l_1},k_{i^*}^{l_2})=1\\ \text{None}, & \text{if } f_e(x^{l_1},k_{i^*}^{l_2})=0 \end{cases} \] with \[ i^*=\arg\max_i P\big(f_e(x^{l_1},k_i^{l_2})=1\big). \] The architecture has two stages. First, a multilingual retriever, implemented with a sentence-transformer/XLM-R-based encoder and a binary classification head, maps the query and knowledge items into a shared multilingual embedding space. Second, the retrieved fact is concatenated with the user prompt, optionally together with bilingual few-shot demonstrations, so that the LLM performs in-context editing rather than parameter modification. Zero-shot ReMaKE models \[ P(y^{l_1}\mid x^{l_1},k^{l_2}), \] whereas few-shot ReMaKE models \[ P(y^{l_1}\mid x^{l_1},k^{l_2},S). \] The bilingual examples are selected by unsupervised cosine-similarity retrieval with **all-MiniLM-L6-v2**. The evaluation uses **MzsRE**, a multilingual version of zsRE covering **12 languages**—English, Czech, German, Dutch, Spanish, French, Portuguese, Russian, Thai, Turkish, Vietnamese, and Chinese. After deduplication, the dataset contains **743 unique items** and includes paraphrased, unrelated, and portability questions [2312.13040]. The four metrics are **Reliability**, **Generality**, **Locality**, and **Portability**. The retriever attains **over 90% accuracy in all languages** and around **97–100%** in many language pairs. Empirically, **ReMaKE-few-bi** is the strongest overall variant, while **ReMaKE-zero** is usually best for portability because it introduces less prompt interference. On English-edit to other-language testing, the reliability gains over IKE range from about **+24.76** for Czech to **+58.72** for Russian. At the same time, portability remains low for all methods, and languages such as **Chinese, Russian, Thai, and Turkish** are harder. A plausible implication is that ReMaKE’s main contribution is not universal reasoning transfer but scalable cross-lingual factual redirection with strong locality. ## 3. ReMake as local video regeneration with motion and content control In video generation, the term is operationalized by **ReVideo: Remake a Video with Motion and Content Control**, which defines local video editing as joint control over *what appears* and *how it moves* [2405.13865]. The task takes as input a video \(\mathbf{V}\in\mathbb{R}^{N\times 3\times W\times H}\) and an editing mask \(\mathbf{M}\in\mathbb{R}^{1\times 1\times W\times H}\). The unedited content is \[ \mathbf{c}_{con}=\mathbf{V}\cdot \mathbf{M}, \] content editing is anchored by a modified first frame \[ \mathbf{c}_{ref}\in\mathbb{R}^{3\times W\times H}, \] and motion control is given by trajectory maps \[ \mathbf{c}_{mot}\in\mathbb{R}^{N\times 2\times W\times H}. \] The generator is Stable Video Diffusion, with denoising written as \[ \hat{\mathbf{z}_0}=\Phi_\theta(\mathbf{z}_t,t,\mathbf{c}_I), \] under EDM preconditioning. The core difficulty is a coupling imbalance between dense unedited content and sparse trajectory signals. ReVideo addresses this through a **three-stage training strategy**. **Motion prior training** first learns trajectory control in isolation. **Decoupling training** then constructs masked composites from two different videos, \[ \mathbf{V}=\mathbf{V}_1\cdot \mathbf{M}+\mathbf{V}_2\cdot (1-\mathbf{M}), \] so that motion inside the edit region cannot be inferred from the preserved context. **Deblocking training** finally restores coherence while updating only the key and value embeddings \(\mathbf{W}_k\) and \(\mathbf{W}_v\) in temporal self-attention layers. The remaining fusion problem is handled by the **spatiotemporal adaptive fusion module (SAFM)**: \[ \mathbf{f}_c=E_c(\mathbf{c}_{con})\cdot \mathbf{\Gamma}+E_m(\mathbf{c}_{mot})\cdot (1-\mathbf{\Gamma}),\qquad \mathbf{\Gamma}=\mathcal{H}(\mathbf{M},t). \] This means that content and motion conditions are not added with a fixed weight; the mixture varies across space and denoising time. On a test set of **16 videos at \(720\times1280\)**, ReVideo is compared with InsV2V, AnyV2V, and Pika. The reported human-preference results are **59.1%** overall preference and **67.0%** editing-target satisfaction for ReVideo, versus **27.9%/23.9%** for Pika, **2.8%/4.0%** for AnyV2V, and **10.2%/5.1%** for InsV2V [2405.13865]. Automatic metrics include **CLIP text-alignment 0.2304**, **temporal consistency 0.9864**, and **PSNR 32.85**. The three canonical applications are local content change with fixed motion, motion change with fixed content, and simultaneous change of both content and motion, and the system also extends to multi-area editing without specific retraining. In this usage, “remake” denotes localized, controllable regeneration rather than whole-video re-synthesis. ## 4. Visual remaking as normalization and long-domain translation A second visual sense of ReMake concerns reconstruction across substantial nuisance factors or domain gaps. **BareSkinNet** is directly relevant to re-makeup because it removes both makeup and illumination while reconstructing a normalized 3D face representation [2209.09029]. The 3D morphable model is \[ S=\bar{S}+B_{id}\alpha+B_{exp}\beta,\qquad A=\bar{A}+B_a\delta, \] with full coefficient vector \[ C(\alpha,\beta,\delta,R,t)\in\mathbb{R}^{298}, \] and lighting \[ \gamma\in\mathbb{R}^{27} \] modeled by spherical harmonics. A pretrained teacher network \(F\) and trainable student network \(E\) provide reconstruction, while a U-Net \(G\) predicts a bare-skin image. The de-makeup loss is \[ L_{DD}=w_1L_{photo}+w_2L_{GAN}+w_3L_{LPIPS}, \] and the full training objective is \[ L_{BSN}=L_{DD}+L_{FR}, \] with \[ L_{FR}=w_{coeff}L_{coeff}+w_{land}L_{land}+w_{diff}L_{diff}+w_{light}L_{light}+w_{reg}L_{reg}. \] The key mechanism is that the rendered no-light 3DMM is forced to match the bare-skin prediction, while the rendered lit 3DMM matches the reference image. Once normalized, the pipeline infers **diffuse, normal, roughness, and specular** maps by unwarping the bare-skin image, 3DMM diffuse, 3DMM normal, and 3DMM geometry into UV space. The paper states that this is done with a **pix2pixHD-style framework**, followed by **SRGAN** upsampling to **4K** [2209.09029]. The system therefore supports makeup removal, makeup extraction, makeup transfer or re-makeup, relighting, and **3D makeup avatar creation**. It reportedly outperforms BeautyGAN, PSGAN, SCGAN, EleGANt, and LADN, and yields lower RMSE with higher PSNR and SSIM when used as preprocessing for texture reconstruction. A more radical visual sense appears in **“Jurassic World Remake: Bringing Ancient Fossils Back to Life via Zero-Shot Long Image-to-Image Translation”**, which defines **longI2I** as translation across large domain gaps where new visual features and new geometry must be generated [2308.07316]. The benchmark method **Revive-2I** maps a source image \(x^s\) to latent space by \[ z^s=\mathcal{E}(x^s), \] applies partial forward diffusion at levels from **50%** to **100%**, and decodes under text guidance such as *“a photo of the head of a \ dog”*. On Skull2Dog, the best reported setting is **Revive-2I 95%**, with **FID 143.29**, **KID 0.0345**, **All @1 100**, and **Class @1 92.56**, outperforming CycleGAN, ACL-GAN, DDIB, and Null-text inversion [2308.07316]. The paper’s broader claim is that guided diffusion is necessary for large domain gaps because prior knowledge about the target domain is required. Together, BareSkinNet and Revive-2I show two complementary remake regimes: nuisance-disentangling normalization and zero-shot semantic re-instantiation. ## 5. ReMake as dataset re-curation and targeted selection In data-centric learning, remake denotes deliberate corpus reconstruction rather than example-level editing. **SEED** formulates data selection as a **Weighted Independent Set (WIS)** problem on a similarity graph [2605.15691]. With graph \(G=(V,E)\) and node weights \(\{w_v\}_{v\in V}\), \[ S^*=\arg\max_{S\subseteq V}\sum_{v\in S}w_v \quad \text{s.t.}\quad (u,v)\notin E,\ \forall u,v\in S. \] Nodes are weighted by trajectory-based influence and edges encode redundancy. The baseline weight for sample \(z_i\) is \[ w_i\coloneqq \max_{z'\in\mathcal{D}_{\text{target}}}\text{Inf}_{\text{traj}}(z_i,z'), \] where the trajectory influence follows TracIn-style gradient alignment. The two main refinements are **node value calibration** and **local scale normalization**. Node calibration restricts influence estimation to a bilateral salient subspace \(\mathcal{C}^*\), defined by channels salient on both training and target gradients, so that influence is recomputed as \[ \text{Inf}_{\text{traj}^*}(z_i,z'_j) =\sum_{t\in\mathcal{T}(z_i)}\eta_t\, \langle \nabla_{\mathcal{C}^*}\mathcal{L}(z_i;\theta_t), \nabla_{\mathcal{C}^*}\mathcal{L}(z'_j;\theta_t)\rangle. \] Local scale normalization estimates density with the \(k\)-th nearest neighbor, \[ \sigma_i\coloneqq \text{Inf}_{\text{train}}(z_i,z_i^{(k)}), \qquad \tau_i\coloneqq \max(\tau,\alpha\cdot \sigma_i), \] and adds edges only when \[ (i,j)\in E^* \iff \text{Inf}_{\text{train}}(z_i,z_j)>\max(\tau_i,\tau_j). \] In practice, the conflict graph is built with **FAISS**, and WIS is solved approximately by a greedy max-heap procedure. The concrete remake product is **Honeybee-Remake-SEED-200K**. Starting from **Honeybee-1M**, annotations are refreshed with **Doubao-1.6-VL**, SEED is run independently for seven target benchmarks, and examples receive votes \[ v_i=\sum_{b=1}^{7}\mathbf{1}[z_i\in\mathcal{S}_b^*]. \] The top **20%** by vote are retained, yielding roughly **200K** multimodal instruction pairs [2605.15691]. Experimentally, SEED reaches an average score of **57.3** at **5%** selection for instruction tuning, matching full-data training and beating TAROT and Nait by about **1.1 points**; at **1%** it reaches **58.2**, exceeding MoNA by **1.0 point**. In visual instruction tuning it attains **65.4**, above the full-data baseline **65.1** and above DataTailor by **1.2 points**. In semantic segmentation it reaches **61.41 mIoU** at **20%**, which is **3.01 points above full-data training**. A highlighted proxy setting uses **Qwen3-1.7B** to select **2.5%** of the data for **Phi-4-14B**, reducing GPU cost by **2.5×** and saving around **138 A100 GPU hours**. Here, remake is best understood as principled re-curation: the corpus is rebuilt to optimize quality, diversity, and downstream utility simultaneously. ## 6. Operational, security, and interactive-media workflows Outside generative modeling, remake often denotes a decision or threat model inside a larger pipeline. In **SlideNet**, automatic quality assessment of Gram-stained direct smear slides is explicitly linked to remake decisions [1803.07240]. If staining quality is low, the system can **send a request to the automatic slide preparation system to remake the slide**; if damage is severe, it can instead trigger manual microscope reading. The pipeline first performs dense region classification with a MobileNet-based CNN and then derives a **region distribution histogram** for slide-level judgments. The eight region labels are **Crystalized**, **Damaged**, **Dense**, **Dirt**, **Edge**, **Empty**, **EpiOnly**, and **LeukOnly**. A region information-density score is defined by \[ I(r)=\lambda_1\cdot \mathcal{S}(l_1(r))+\lambda_2\cdot \mathcal{S}(l_2(r)), \] with \(\lambda_1=0.8\), \(\lambda_2=0.2\), and \(\mathcal{S}(l)=-x^2+\lambda_3\), \(\lambda_3=256\). On patch classification, **SlideNet-224** reaches **86.8%** accuracy versus **45.3%** for HOG+SVM and **51.2%** for colorLBP+SVM, and slide-level agreement with expert opinions is **0.778**. In this setting, remake is a quality-gated remediation step. In face anti-spoofing, **CLIPC8** treats **screen remake** or replay as one element of a richer eight-class taxonomy [2311.17583]. The model uses a pretrained **CLIP ViT-L/14** backbone, with image-text contrastive alignment based on cosine similarity \[ S=\cos(I,T)=\frac{I\cdot T}{\|I\|_2\|T\|_2}, \] followed by softmax over eight prompts. Category **(f) Recaptured Photo** explicitly includes personal photos or ID photos displayed on a **mobile phone** or **computer screen**, with cues such as **screen borders**, **moiré patterns**, and **partial information of the hands**. Zero-shot on **Replay-Attack**, CLIPC8 reports **Real 89.03**, **Fake 85.12**, and **ACER 0.595%**; after broader fine-tuning it reports **99.00** real and **99.81** fake on the same benchmark [2311.17583]. The relevance of remake here is adversarial rather than constructive: the system must detect recaptured or replayed media that visually re-present a prior face sample. A more narrative use appears in **“What if Red Can Talk?”**, which studies dynamic filler dialogue generation with knowledge graphs and GPT-4 in **Final Fantasy VII Remake** and Pokémon [2407.20382]. For FFVIIR, the experiments use **three bosses**, **13 battle situations**, and **65 GPT-4 responses**. The output is grounded in RDF-style triples for Cloud and boss states, but the paper notes that Cloud often sounds **overly positive** and that responses are frequently driven more by immediate situation than by explicit knowledge-graph facts. This usage does not define a ReMake framework, but it shows how a title-bearing cultural artifact can become a testbed for constrained regeneration of dialogue. ## 7. Scientific and mathematical extensions of the remake concept In several papers, remake is neither a framework name nor an editing task, but a mechanistic or conceptual descriptor. In granular mechanics, dynamic softening is attributed not primarily to microslip but to **break and remake of contacts** under AC-driven rearrangement [1706.05898]. The DEM study defines inter-particle forces with Hertzian normal contact, tangential frictional contact, viscous damping, and rolling resistance, and shows that resonance frequencies shift downward as drive amplitude increases. The **Slipping Contact Ratio** changes little when friction is varied, whereas the **coordination number** \(Z_c\) decreases at resonance and strongly tracks softening. The paper’s interpretation is that extended structural change—contact breakage and remake—reduces connectivity and hence stiffness. In political modeling, **“After 2018 Bolsonaro victory, is a 2022 remake feasible?”** uses remake in the sense of a rematch [2112.13713]. The paper models Brazilian voting trends with left and right groups containing sensitives, inflexibles, and radicals, plus an optional populist group \(p\). In the populist scenario, \[ p=f\frac{\alpha}{1-2\alpha} \] for \(\alpha<0.5\), and when \(\alpha\ge 0.5\) the model yields \(l=r=0\) and \[ p=1-4f. \] With \(f=0.04\), this gives \(p=0.84\). The paper concludes that a 2022 Lula/Bolsonaro remake is feasible in either scenario: Lula can win in the low-populism regime if \(k\) shifts sufficiently, whereas Bolsonaro becomes favored if the populist coupling crosses the threshold near \(\alpha=0.5\). In analysis, **“A Remake of Bourgain-Brezis-Mironescu Characterization of Sobolev Spaces”** generalizes the BBM limit formula from classical fractional kernels to concentrated \(p\)-Lévy kernels [2008.07631]. The paper studies nonlocal energies \[ E_i(u)=\iint_{\mathbb{R}^d\times\mathbb{R}^d}|u(x)-u(y)|^p\,v(x-y)\,dx\,dy, \] proves that finiteness of the liminf characterizes \(W^{1,p}\) for \(1